{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 7.3591002044989775, "eval_steps": 500, "global_step": 9000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "entropy": 10.692000198364259, "epoch": 0.00408997955010225, "grad_norm": 12.6875, "learning_rate": 2e-06, "loss": 11.0015, "mean_token_accuracy": 2.4464876332785934e-05, "num_tokens": 81527.0, "step": 5 }, { "entropy": 10.691902351379394, "epoch": 0.0081799591002045, "grad_norm": 12.5, "learning_rate": 4.5e-06, "loss": 10.9107, "mean_token_accuracy": 0.0001590834290254861, "num_tokens": 162815.0, "step": 10 }, { "entropy": 10.690652847290039, "epoch": 0.012269938650306749, "grad_norm": 8.0, "learning_rate": 7e-06, "loss": 10.5862, "mean_token_accuracy": 0.021183770173229276, "num_tokens": 244175.0, "step": 15 }, { "entropy": 10.67988691329956, "epoch": 0.016359918200409, "grad_norm": 4.46875, "learning_rate": 9.5e-06, "loss": 10.2158, "mean_token_accuracy": 0.03468580432236194, "num_tokens": 324928.0, "step": 20 }, { "entropy": 10.636291885375977, "epoch": 0.02044989775051125, "grad_norm": 2.796875, "learning_rate": 1.2e-05, "loss": 9.8963, "mean_token_accuracy": 0.039024268835783006, "num_tokens": 406579.0, "step": 25 }, { "entropy": 10.604203414916991, "epoch": 0.024539877300613498, "grad_norm": 2.4375, "learning_rate": 1.4500000000000002e-05, "loss": 9.7363, "mean_token_accuracy": 0.04346781931817532, "num_tokens": 487674.0, "step": 30 }, { "entropy": 10.598912239074707, "epoch": 0.028629856850715747, "grad_norm": 2.296875, "learning_rate": 1.7000000000000003e-05, "loss": 9.6794, "mean_token_accuracy": 0.04639807604253292, "num_tokens": 569303.0, "step": 35 }, { "entropy": 10.597527599334716, "epoch": 0.032719836400818, "grad_norm": 2.375, "learning_rate": 1.95e-05, "loss": 9.6149, "mean_token_accuracy": 0.05854539461433887, "num_tokens": 650813.0, "step": 40 }, { "entropy": 10.537779712677002, "epoch": 0.03680981595092025, "grad_norm": 2.484375, "learning_rate": 2.2e-05, "loss": 9.493, "mean_token_accuracy": 0.07012398801743984, "num_tokens": 732280.0, "step": 45 }, { "entropy": 10.376054859161377, "epoch": 0.0408997955010225, "grad_norm": 2.4375, "learning_rate": 2.4500000000000003e-05, "loss": 9.3922, "mean_token_accuracy": 0.0757705345749855, "num_tokens": 813564.0, "step": 50 }, { "entropy": 10.325862598419189, "epoch": 0.044989775051124746, "grad_norm": 2.1875, "learning_rate": 2.7e-05, "loss": 9.3308, "mean_token_accuracy": 0.07539180815219879, "num_tokens": 895017.0, "step": 55 }, { "entropy": 10.389144897460938, "epoch": 0.049079754601226995, "grad_norm": 2.140625, "learning_rate": 2.95e-05, "loss": 9.2897, "mean_token_accuracy": 0.07736093625426292, "num_tokens": 976507.0, "step": 60 }, { "entropy": 10.31296672821045, "epoch": 0.053169734151329244, "grad_norm": 1.921875, "learning_rate": 3.2e-05, "loss": 9.1381, "mean_token_accuracy": 0.08499588370323181, "num_tokens": 1057953.0, "step": 65 }, { "entropy": 10.195121669769287, "epoch": 0.05725971370143149, "grad_norm": 2.046875, "learning_rate": 3.4500000000000005e-05, "loss": 8.9981, "mean_token_accuracy": 0.0928544394671917, "num_tokens": 1138725.0, "step": 70 }, { "entropy": 10.178840732574463, "epoch": 0.06134969325153374, "grad_norm": 2.421875, "learning_rate": 3.7e-05, "loss": 8.9206, "mean_token_accuracy": 0.09163383767008781, "num_tokens": 1219783.0, "step": 75 }, { "entropy": 10.1720477104187, "epoch": 0.065439672801636, "grad_norm": 1.9921875, "learning_rate": 3.95e-05, "loss": 8.7942, "mean_token_accuracy": 0.09474985226988793, "num_tokens": 1300892.0, "step": 80 }, { "entropy": 10.115754795074462, "epoch": 0.06952965235173825, "grad_norm": 1.84375, "learning_rate": 4.2000000000000004e-05, "loss": 8.6972, "mean_token_accuracy": 0.09509614557027816, "num_tokens": 1382196.0, "step": 85 }, { "entropy": 10.115432357788086, "epoch": 0.0736196319018405, "grad_norm": 2.046875, "learning_rate": 4.45e-05, "loss": 8.5859, "mean_token_accuracy": 0.09679216295480728, "num_tokens": 1463433.0, "step": 90 }, { "entropy": 10.048014736175537, "epoch": 0.07770961145194274, "grad_norm": 1.765625, "learning_rate": 4.7000000000000004e-05, "loss": 8.4925, "mean_token_accuracy": 0.09578462094068527, "num_tokens": 1544967.0, "step": 95 }, { "entropy": 10.00209379196167, "epoch": 0.081799591002045, "grad_norm": 1.8515625, "learning_rate": 4.9500000000000004e-05, "loss": 8.3486, "mean_token_accuracy": 0.10033286660909653, "num_tokens": 1626565.0, "step": 100 }, { "entropy": 9.9131498336792, "epoch": 0.08588957055214724, "grad_norm": 1.78125, "learning_rate": 5.2e-05, "loss": 8.2137, "mean_token_accuracy": 0.10160388499498367, "num_tokens": 1707817.0, "step": 105 }, { "entropy": 9.82022647857666, "epoch": 0.08997955010224949, "grad_norm": 1.6484375, "learning_rate": 5.45e-05, "loss": 8.0758, "mean_token_accuracy": 0.10720168352127075, "num_tokens": 1789495.0, "step": 110 }, { "entropy": 9.724441719055175, "epoch": 0.09406952965235174, "grad_norm": 1.4140625, "learning_rate": 5.7e-05, "loss": 7.9726, "mean_token_accuracy": 0.10897153615951538, "num_tokens": 1871084.0, "step": 115 }, { "entropy": 9.586730098724365, "epoch": 0.09815950920245399, "grad_norm": 1.3203125, "learning_rate": 5.9499999999999996e-05, "loss": 7.8413, "mean_token_accuracy": 0.10868334248661995, "num_tokens": 1952298.0, "step": 120 }, { "entropy": 9.416004085540772, "epoch": 0.10224948875255624, "grad_norm": 1.4765625, "learning_rate": 6.2e-05, "loss": 7.6389, "mean_token_accuracy": 0.11382156684994697, "num_tokens": 2033946.0, "step": 125 }, { "entropy": 9.186109161376953, "epoch": 0.10633946830265849, "grad_norm": 1.171875, "learning_rate": 6.450000000000001e-05, "loss": 7.5292, "mean_token_accuracy": 0.11576363742351532, "num_tokens": 2115355.0, "step": 130 }, { "entropy": 8.975370025634765, "epoch": 0.11042944785276074, "grad_norm": 1.2421875, "learning_rate": 6.7e-05, "loss": 7.4296, "mean_token_accuracy": 0.11505270451307296, "num_tokens": 2195805.0, "step": 135 }, { "entropy": 8.75938024520874, "epoch": 0.11451942740286299, "grad_norm": 1.1640625, "learning_rate": 6.950000000000001e-05, "loss": 7.3127, "mean_token_accuracy": 0.11694714352488518, "num_tokens": 2277317.0, "step": 140 }, { "entropy": 8.507802486419678, "epoch": 0.11860940695296524, "grad_norm": 1.0703125, "learning_rate": 7.2e-05, "loss": 7.2511, "mean_token_accuracy": 0.1210237443447113, "num_tokens": 2358952.0, "step": 145 }, { "entropy": 8.29345064163208, "epoch": 0.12269938650306748, "grad_norm": 1.078125, "learning_rate": 7.45e-05, "loss": 7.1339, "mean_token_accuracy": 0.12062957882881165, "num_tokens": 2440370.0, "step": 150 }, { "entropy": 8.062093591690063, "epoch": 0.12678936605316973, "grad_norm": 0.92578125, "learning_rate": 7.7e-05, "loss": 7.1056, "mean_token_accuracy": 0.1271037481725216, "num_tokens": 2521468.0, "step": 155 }, { "entropy": 7.970836448669433, "epoch": 0.130879345603272, "grad_norm": 1.0625, "learning_rate": 7.950000000000001e-05, "loss": 7.0757, "mean_token_accuracy": 0.1223635084927082, "num_tokens": 2601986.0, "step": 160 }, { "entropy": 7.8029450416564945, "epoch": 0.13496932515337423, "grad_norm": 0.96484375, "learning_rate": 8.2e-05, "loss": 6.9867, "mean_token_accuracy": 0.12624209150671958, "num_tokens": 2683683.0, "step": 165 }, { "entropy": 7.705253219604492, "epoch": 0.1390593047034765, "grad_norm": 1.4609375, "learning_rate": 8.450000000000001e-05, "loss": 6.985, "mean_token_accuracy": 0.12793700769543648, "num_tokens": 2765248.0, "step": 170 }, { "entropy": 7.637160062789917, "epoch": 0.14314928425357873, "grad_norm": 1.0703125, "learning_rate": 8.7e-05, "loss": 6.9519, "mean_token_accuracy": 0.12585932165384292, "num_tokens": 2846905.0, "step": 175 }, { "entropy": 7.6360515594482425, "epoch": 0.147239263803681, "grad_norm": 1.03125, "learning_rate": 8.95e-05, "loss": 6.953, "mean_token_accuracy": 0.1275423213839531, "num_tokens": 2927747.0, "step": 180 }, { "entropy": 7.516210556030273, "epoch": 0.15132924335378323, "grad_norm": 0.828125, "learning_rate": 9.2e-05, "loss": 6.8677, "mean_token_accuracy": 0.13259214907884598, "num_tokens": 3008816.0, "step": 185 }, { "entropy": 7.476730108261108, "epoch": 0.1554192229038855, "grad_norm": 1.140625, "learning_rate": 9.45e-05, "loss": 6.8269, "mean_token_accuracy": 0.13380919247865677, "num_tokens": 3090309.0, "step": 190 }, { "entropy": 7.442372560501099, "epoch": 0.15950920245398773, "grad_norm": 0.86328125, "learning_rate": 9.7e-05, "loss": 6.8692, "mean_token_accuracy": 0.13351524621248245, "num_tokens": 3171767.0, "step": 195 }, { "entropy": 7.416834020614624, "epoch": 0.16359918200409, "grad_norm": 0.90625, "learning_rate": 9.95e-05, "loss": 6.8286, "mean_token_accuracy": 0.13438718765974045, "num_tokens": 3253322.0, "step": 200 }, { "entropy": 7.4071942329406735, "epoch": 0.16768916155419222, "grad_norm": 1.109375, "learning_rate": 0.000102, "loss": 6.8339, "mean_token_accuracy": 0.13587063997983934, "num_tokens": 3334617.0, "step": 205 }, { "entropy": 7.334885931015014, "epoch": 0.17177914110429449, "grad_norm": 1.0625, "learning_rate": 0.00010449999999999999, "loss": 6.8066, "mean_token_accuracy": 0.1350346751511097, "num_tokens": 3416224.0, "step": 210 }, { "entropy": 7.294337224960327, "epoch": 0.17586912065439672, "grad_norm": 0.859375, "learning_rate": 0.000107, "loss": 6.7851, "mean_token_accuracy": 0.13475466817617415, "num_tokens": 3497574.0, "step": 215 }, { "entropy": 7.308300495147705, "epoch": 0.17995910020449898, "grad_norm": 1.0078125, "learning_rate": 0.0001095, "loss": 6.7822, "mean_token_accuracy": 0.13686141669750213, "num_tokens": 3578525.0, "step": 220 }, { "entropy": 7.253078174591065, "epoch": 0.18404907975460122, "grad_norm": 0.84375, "learning_rate": 0.000112, "loss": 6.7709, "mean_token_accuracy": 0.1378871127963066, "num_tokens": 3660181.0, "step": 225 }, { "entropy": 7.265198373794556, "epoch": 0.18813905930470348, "grad_norm": 0.765625, "learning_rate": 0.0001145, "loss": 6.7519, "mean_token_accuracy": 0.13652033507823944, "num_tokens": 3741809.0, "step": 230 }, { "entropy": 7.23785982131958, "epoch": 0.19222903885480572, "grad_norm": 0.98046875, "learning_rate": 0.00011700000000000001, "loss": 6.7088, "mean_token_accuracy": 0.13769972324371338, "num_tokens": 3823389.0, "step": 235 }, { "entropy": 7.1915041446685795, "epoch": 0.19631901840490798, "grad_norm": 0.9140625, "learning_rate": 0.00011949999999999999, "loss": 6.7275, "mean_token_accuracy": 0.13537857830524444, "num_tokens": 3904714.0, "step": 240 }, { "entropy": 7.14861159324646, "epoch": 0.20040899795501022, "grad_norm": 1.046875, "learning_rate": 0.000122, "loss": 6.6742, "mean_token_accuracy": 0.1384981632232666, "num_tokens": 3986263.0, "step": 245 }, { "entropy": 7.149033546447754, "epoch": 0.20449897750511248, "grad_norm": 0.9375, "learning_rate": 0.0001245, "loss": 6.6775, "mean_token_accuracy": 0.13888683989644052, "num_tokens": 4067896.0, "step": 250 }, { "entropy": 7.114649248123169, "epoch": 0.2085889570552147, "grad_norm": 0.9375, "learning_rate": 0.000127, "loss": 6.6754, "mean_token_accuracy": 0.1410924032330513, "num_tokens": 4149422.0, "step": 255 }, { "entropy": 7.072041988372803, "epoch": 0.21267893660531698, "grad_norm": 0.84375, "learning_rate": 0.0001295, "loss": 6.5748, "mean_token_accuracy": 0.14269786030054094, "num_tokens": 4230807.0, "step": 260 }, { "entropy": 7.01435227394104, "epoch": 0.2167689161554192, "grad_norm": 1.0546875, "learning_rate": 0.000132, "loss": 6.6012, "mean_token_accuracy": 0.14459822773933412, "num_tokens": 4312299.0, "step": 265 }, { "entropy": 7.081182289123535, "epoch": 0.22085889570552147, "grad_norm": 0.8984375, "learning_rate": 0.00013450000000000002, "loss": 6.6316, "mean_token_accuracy": 0.1434426262974739, "num_tokens": 4393563.0, "step": 270 }, { "entropy": 7.031969118118286, "epoch": 0.2249488752556237, "grad_norm": 0.8046875, "learning_rate": 0.00013700000000000002, "loss": 6.6549, "mean_token_accuracy": 0.13694730624556542, "num_tokens": 4474907.0, "step": 275 }, { "entropy": 7.04373688697815, "epoch": 0.22903885480572597, "grad_norm": 0.87890625, "learning_rate": 0.0001395, "loss": 6.5587, "mean_token_accuracy": 0.14282752722501754, "num_tokens": 4556026.0, "step": 280 }, { "entropy": 7.012068223953247, "epoch": 0.2331288343558282, "grad_norm": 1.28125, "learning_rate": 0.00014199999999999998, "loss": 6.5625, "mean_token_accuracy": 0.142834110558033, "num_tokens": 4637280.0, "step": 285 }, { "entropy": 6.986003494262695, "epoch": 0.23721881390593047, "grad_norm": 0.8671875, "learning_rate": 0.0001445, "loss": 6.5745, "mean_token_accuracy": 0.14551258087158203, "num_tokens": 4718900.0, "step": 290 }, { "entropy": 6.98618369102478, "epoch": 0.24130879345603273, "grad_norm": 0.90625, "learning_rate": 0.000147, "loss": 6.5734, "mean_token_accuracy": 0.142940554022789, "num_tokens": 4800507.0, "step": 295 }, { "entropy": 6.899418592453003, "epoch": 0.24539877300613497, "grad_norm": 0.859375, "learning_rate": 0.0001495, "loss": 6.5152, "mean_token_accuracy": 0.14577620029449462, "num_tokens": 4881971.0, "step": 300 }, { "entropy": 6.931239128112793, "epoch": 0.24948875255623723, "grad_norm": 0.91015625, "learning_rate": 0.000152, "loss": 6.5192, "mean_token_accuracy": 0.1479423984885216, "num_tokens": 4963516.0, "step": 305 }, { "entropy": 6.870612716674804, "epoch": 0.25357873210633947, "grad_norm": 0.96875, "learning_rate": 0.00015450000000000001, "loss": 6.4973, "mean_token_accuracy": 0.14800097942352294, "num_tokens": 5045190.0, "step": 310 }, { "entropy": 6.911778402328491, "epoch": 0.25766871165644173, "grad_norm": 0.9375, "learning_rate": 0.000157, "loss": 6.5103, "mean_token_accuracy": 0.14730968773365022, "num_tokens": 5126730.0, "step": 315 }, { "entropy": 6.862739562988281, "epoch": 0.261758691206544, "grad_norm": 0.86328125, "learning_rate": 0.0001595, "loss": 6.5129, "mean_token_accuracy": 0.14664446264505387, "num_tokens": 5207888.0, "step": 320 }, { "entropy": 6.849811601638794, "epoch": 0.2658486707566462, "grad_norm": 0.81640625, "learning_rate": 0.000162, "loss": 6.437, "mean_token_accuracy": 0.1522682175040245, "num_tokens": 5289484.0, "step": 325 }, { "entropy": 6.818774795532226, "epoch": 0.26993865030674846, "grad_norm": 0.75, "learning_rate": 0.00016450000000000001, "loss": 6.405, "mean_token_accuracy": 0.15261074155569077, "num_tokens": 5371086.0, "step": 330 }, { "entropy": 6.868287801742554, "epoch": 0.2740286298568507, "grad_norm": 0.77734375, "learning_rate": 0.00016700000000000002, "loss": 6.5087, "mean_token_accuracy": 0.1447703406214714, "num_tokens": 5452574.0, "step": 335 }, { "entropy": 6.7703859329223635, "epoch": 0.278118609406953, "grad_norm": 0.89453125, "learning_rate": 0.00016950000000000003, "loss": 6.4103, "mean_token_accuracy": 0.1545592203736305, "num_tokens": 5534091.0, "step": 340 }, { "entropy": 6.787615823745727, "epoch": 0.2822085889570552, "grad_norm": 0.87109375, "learning_rate": 0.00017199999999999998, "loss": 6.4559, "mean_token_accuracy": 0.14936767816543578, "num_tokens": 5614709.0, "step": 345 }, { "entropy": 6.780069017410279, "epoch": 0.28629856850715746, "grad_norm": 0.96484375, "learning_rate": 0.00017449999999999999, "loss": 6.3975, "mean_token_accuracy": 0.15142018646001815, "num_tokens": 5695945.0, "step": 350 }, { "entropy": 6.7227113246917725, "epoch": 0.2903885480572597, "grad_norm": 0.890625, "learning_rate": 0.000177, "loss": 6.3873, "mean_token_accuracy": 0.15569809824228287, "num_tokens": 5776894.0, "step": 355 }, { "entropy": 6.681255960464478, "epoch": 0.294478527607362, "grad_norm": 0.83984375, "learning_rate": 0.0001795, "loss": 6.3418, "mean_token_accuracy": 0.1554790660738945, "num_tokens": 5858426.0, "step": 360 }, { "entropy": 6.75174880027771, "epoch": 0.2985685071574642, "grad_norm": 0.7890625, "learning_rate": 0.000182, "loss": 6.435, "mean_token_accuracy": 0.1517605572938919, "num_tokens": 5939990.0, "step": 365 }, { "entropy": 6.69475622177124, "epoch": 0.30265848670756645, "grad_norm": 0.7578125, "learning_rate": 0.0001845, "loss": 6.3261, "mean_token_accuracy": 0.15980565249919892, "num_tokens": 6021209.0, "step": 370 }, { "entropy": 6.72758207321167, "epoch": 0.3067484662576687, "grad_norm": 0.89453125, "learning_rate": 0.000187, "loss": 6.3882, "mean_token_accuracy": 0.15352147072553635, "num_tokens": 6102085.0, "step": 375 }, { "entropy": 6.668760585784912, "epoch": 0.310838445807771, "grad_norm": 0.90234375, "learning_rate": 0.0001895, "loss": 6.3174, "mean_token_accuracy": 0.1569987043738365, "num_tokens": 6183631.0, "step": 380 }, { "entropy": 6.686910724639892, "epoch": 0.3149284253578732, "grad_norm": 0.72265625, "learning_rate": 0.000192, "loss": 6.3469, "mean_token_accuracy": 0.15672450959682466, "num_tokens": 6264174.0, "step": 385 }, { "entropy": 6.64973726272583, "epoch": 0.31901840490797545, "grad_norm": 0.87109375, "learning_rate": 0.0001945, "loss": 6.2917, "mean_token_accuracy": 0.1605835884809494, "num_tokens": 6345616.0, "step": 390 }, { "entropy": 6.6427412033081055, "epoch": 0.3231083844580777, "grad_norm": 0.81640625, "learning_rate": 0.00019700000000000002, "loss": 6.2962, "mean_token_accuracy": 0.15768682956695557, "num_tokens": 6426371.0, "step": 395 }, { "entropy": 6.668876314163208, "epoch": 0.32719836400818, "grad_norm": 0.95703125, "learning_rate": 0.00019950000000000002, "loss": 6.349, "mean_token_accuracy": 0.15605812668800353, "num_tokens": 6508017.0, "step": 400 }, { "entropy": 6.626045894622803, "epoch": 0.3312883435582822, "grad_norm": 1.0234375, "learning_rate": 0.000202, "loss": 6.3398, "mean_token_accuracy": 0.15648607313632965, "num_tokens": 6589627.0, "step": 405 }, { "entropy": 6.614638900756836, "epoch": 0.33537832310838445, "grad_norm": 0.90625, "learning_rate": 0.00020449999999999998, "loss": 6.2965, "mean_token_accuracy": 0.16072938144207, "num_tokens": 6670965.0, "step": 410 }, { "entropy": 6.662922954559326, "epoch": 0.3394683026584867, "grad_norm": 0.92578125, "learning_rate": 0.000207, "loss": 6.3392, "mean_token_accuracy": 0.15543674528598786, "num_tokens": 6752603.0, "step": 415 }, { "entropy": 6.557052993774414, "epoch": 0.34355828220858897, "grad_norm": 0.89453125, "learning_rate": 0.0002095, "loss": 6.2425, "mean_token_accuracy": 0.1611531674861908, "num_tokens": 6833529.0, "step": 420 }, { "entropy": 6.608387994766235, "epoch": 0.3476482617586912, "grad_norm": 0.828125, "learning_rate": 0.000212, "loss": 6.2616, "mean_token_accuracy": 0.1630495846271515, "num_tokens": 6915141.0, "step": 425 }, { "entropy": 6.538517665863037, "epoch": 0.35173824130879344, "grad_norm": 0.81640625, "learning_rate": 0.0002145, "loss": 6.2681, "mean_token_accuracy": 0.158408260345459, "num_tokens": 6996145.0, "step": 430 }, { "entropy": 6.586181640625, "epoch": 0.3558282208588957, "grad_norm": 0.74609375, "learning_rate": 0.00021700000000000002, "loss": 6.2616, "mean_token_accuracy": 0.1599885493516922, "num_tokens": 7077692.0, "step": 435 }, { "entropy": 6.546446752548218, "epoch": 0.35991820040899797, "grad_norm": 0.8125, "learning_rate": 0.0002195, "loss": 6.2648, "mean_token_accuracy": 0.16150236129760742, "num_tokens": 7159364.0, "step": 440 }, { "entropy": 6.547263431549072, "epoch": 0.36400817995910023, "grad_norm": 0.76171875, "learning_rate": 0.000222, "loss": 6.23, "mean_token_accuracy": 0.16269229352474213, "num_tokens": 7240330.0, "step": 445 }, { "entropy": 6.473912286758423, "epoch": 0.36809815950920244, "grad_norm": 0.87890625, "learning_rate": 0.0002245, "loss": 6.1576, "mean_token_accuracy": 0.16461296379566193, "num_tokens": 7320583.0, "step": 450 }, { "entropy": 6.49521541595459, "epoch": 0.3721881390593047, "grad_norm": 0.75, "learning_rate": 0.00022700000000000002, "loss": 6.2006, "mean_token_accuracy": 0.16421556025743483, "num_tokens": 7401685.0, "step": 455 }, { "entropy": 6.5050506591796875, "epoch": 0.37627811860940696, "grad_norm": 0.78515625, "learning_rate": 0.00022950000000000002, "loss": 6.1459, "mean_token_accuracy": 0.1676486998796463, "num_tokens": 7482873.0, "step": 460 }, { "entropy": 6.463202667236328, "epoch": 0.3803680981595092, "grad_norm": 0.73828125, "learning_rate": 0.00023200000000000003, "loss": 6.157, "mean_token_accuracy": 0.1651605486869812, "num_tokens": 7564543.0, "step": 465 }, { "entropy": 6.4412964344024655, "epoch": 0.38445807770961143, "grad_norm": 0.83203125, "learning_rate": 0.00023449999999999998, "loss": 6.186, "mean_token_accuracy": 0.16442248970270157, "num_tokens": 7645311.0, "step": 470 }, { "entropy": 6.522923421859741, "epoch": 0.3885480572597137, "grad_norm": 0.87109375, "learning_rate": 0.000237, "loss": 6.2154, "mean_token_accuracy": 0.16138581335544586, "num_tokens": 7726050.0, "step": 475 }, { "entropy": 6.4519061088562015, "epoch": 0.39263803680981596, "grad_norm": 0.765625, "learning_rate": 0.0002395, "loss": 6.1928, "mean_token_accuracy": 0.1644761323928833, "num_tokens": 7807452.0, "step": 480 }, { "entropy": 6.45891547203064, "epoch": 0.3967280163599182, "grad_norm": 0.7265625, "learning_rate": 0.000242, "loss": 6.1734, "mean_token_accuracy": 0.16323913782835006, "num_tokens": 7889159.0, "step": 485 }, { "entropy": 6.512608671188355, "epoch": 0.40081799591002043, "grad_norm": 0.765625, "learning_rate": 0.0002445, "loss": 6.224, "mean_token_accuracy": 0.16140950471162796, "num_tokens": 7970803.0, "step": 490 }, { "entropy": 6.389305877685547, "epoch": 0.4049079754601227, "grad_norm": 0.734375, "learning_rate": 0.000247, "loss": 6.0861, "mean_token_accuracy": 0.17114942967891694, "num_tokens": 8051777.0, "step": 495 }, { "entropy": 6.496857643127441, "epoch": 0.40899795501022496, "grad_norm": 0.79296875, "learning_rate": 0.0002495, "loss": 6.1937, "mean_token_accuracy": 0.16533919274806977, "num_tokens": 8133204.0, "step": 500 }, { "epoch": 0.40899795501022496, "eval_entropy": 6.272721568589965, "eval_loss": 6.081885814666748, "eval_mean_token_accuracy": 0.1713385213756193, "eval_num_tokens": 8133204.0, "eval_runtime": 66.7776, "eval_samples_per_second": 61.952, "eval_steps_per_second": 7.757, "step": 500 }, { "entropy": 6.445580768585205, "epoch": 0.4130879345603272, "grad_norm": 0.81640625, "learning_rate": 0.000252, "loss": 6.1888, "mean_token_accuracy": 0.16254500448703765, "num_tokens": 8213845.0, "step": 505 }, { "entropy": 6.478588008880616, "epoch": 0.4171779141104294, "grad_norm": 0.8203125, "learning_rate": 0.0002545, "loss": 6.1542, "mean_token_accuracy": 0.1684673473238945, "num_tokens": 8294824.0, "step": 510 }, { "entropy": 6.444344139099121, "epoch": 0.4212678936605317, "grad_norm": 0.86328125, "learning_rate": 0.000257, "loss": 6.2059, "mean_token_accuracy": 0.16449313759803771, "num_tokens": 8376070.0, "step": 515 }, { "entropy": 6.423702383041382, "epoch": 0.42535787321063395, "grad_norm": 0.71484375, "learning_rate": 0.0002595, "loss": 6.1591, "mean_token_accuracy": 0.16637170016765596, "num_tokens": 8457502.0, "step": 520 }, { "entropy": 6.400311899185181, "epoch": 0.4294478527607362, "grad_norm": 0.70703125, "learning_rate": 0.000262, "loss": 6.1068, "mean_token_accuracy": 0.16918586641550065, "num_tokens": 8538963.0, "step": 525 }, { "entropy": 6.4158618450164795, "epoch": 0.4335378323108384, "grad_norm": 0.78125, "learning_rate": 0.00026450000000000003, "loss": 6.1265, "mean_token_accuracy": 0.16910201609134673, "num_tokens": 8620210.0, "step": 530 }, { "entropy": 6.323219633102417, "epoch": 0.4376278118609407, "grad_norm": 0.859375, "learning_rate": 0.00026700000000000004, "loss": 6.0399, "mean_token_accuracy": 0.16840801686048507, "num_tokens": 8700454.0, "step": 535 }, { "entropy": 6.330011558532715, "epoch": 0.44171779141104295, "grad_norm": 0.86328125, "learning_rate": 0.00026950000000000005, "loss": 6.07, "mean_token_accuracy": 0.16758037060499192, "num_tokens": 8781678.0, "step": 540 }, { "entropy": 6.414210176467895, "epoch": 0.4458077709611452, "grad_norm": 0.8125, "learning_rate": 0.00027200000000000005, "loss": 6.1029, "mean_token_accuracy": 0.1698324516415596, "num_tokens": 8863235.0, "step": 545 }, { "entropy": 6.347384023666382, "epoch": 0.4498977505112474, "grad_norm": 1.0078125, "learning_rate": 0.0002745, "loss": 6.0687, "mean_token_accuracy": 0.16842822581529618, "num_tokens": 8944923.0, "step": 550 }, { "entropy": 6.282917213439942, "epoch": 0.4539877300613497, "grad_norm": 0.90234375, "learning_rate": 0.000277, "loss": 6.048, "mean_token_accuracy": 0.16954710185527802, "num_tokens": 9026323.0, "step": 555 }, { "entropy": 6.293161678314209, "epoch": 0.45807770961145194, "grad_norm": 0.73828125, "learning_rate": 0.0002795, "loss": 6.0083, "mean_token_accuracy": 0.1738037198781967, "num_tokens": 9107139.0, "step": 560 }, { "entropy": 6.243445110321045, "epoch": 0.4621676891615542, "grad_norm": 0.69140625, "learning_rate": 0.00028199999999999997, "loss": 5.9823, "mean_token_accuracy": 0.1747031480073929, "num_tokens": 9188143.0, "step": 565 }, { "entropy": 6.313747406005859, "epoch": 0.4662576687116564, "grad_norm": 0.78515625, "learning_rate": 0.0002845, "loss": 6.0631, "mean_token_accuracy": 0.17030100226402284, "num_tokens": 9269739.0, "step": 570 }, { "entropy": 6.329251480102539, "epoch": 0.4703476482617587, "grad_norm": 0.74609375, "learning_rate": 0.000287, "loss": 6.0279, "mean_token_accuracy": 0.17231380194425583, "num_tokens": 9350856.0, "step": 575 }, { "entropy": 6.268042087554932, "epoch": 0.47443762781186094, "grad_norm": 0.82421875, "learning_rate": 0.0002895, "loss": 6.0068, "mean_token_accuracy": 0.17424249649047852, "num_tokens": 9432095.0, "step": 580 }, { "entropy": 6.245525693893432, "epoch": 0.4785276073619632, "grad_norm": 0.89453125, "learning_rate": 0.000292, "loss": 5.9865, "mean_token_accuracy": 0.17295096665620804, "num_tokens": 9513346.0, "step": 585 }, { "entropy": 6.256681776046753, "epoch": 0.48261758691206547, "grad_norm": 0.79296875, "learning_rate": 0.0002945, "loss": 5.9707, "mean_token_accuracy": 0.1737953931093216, "num_tokens": 9594808.0, "step": 590 }, { "entropy": 6.215074586868286, "epoch": 0.4867075664621677, "grad_norm": 0.7734375, "learning_rate": 0.000297, "loss": 5.9689, "mean_token_accuracy": 0.17679460793733598, "num_tokens": 9676232.0, "step": 595 }, { "entropy": 6.2827376365661625, "epoch": 0.49079754601226994, "grad_norm": 0.859375, "learning_rate": 0.0002995, "loss": 6.0254, "mean_token_accuracy": 0.17187506407499314, "num_tokens": 9757797.0, "step": 600 }, { "entropy": 6.20688943862915, "epoch": 0.4948875255623722, "grad_norm": 0.95703125, "learning_rate": 0.000302, "loss": 5.9507, "mean_token_accuracy": 0.17501263171434403, "num_tokens": 9839392.0, "step": 605 }, { "entropy": 6.203517961502075, "epoch": 0.49897750511247446, "grad_norm": 0.7578125, "learning_rate": 0.0003045, "loss": 5.9512, "mean_token_accuracy": 0.17712361216545106, "num_tokens": 9919968.0, "step": 610 }, { "entropy": 6.228668069839477, "epoch": 0.5030674846625767, "grad_norm": 0.8046875, "learning_rate": 0.000307, "loss": 5.9559, "mean_token_accuracy": 0.17686514556407928, "num_tokens": 10000990.0, "step": 615 }, { "entropy": 6.225601863861084, "epoch": 0.5071574642126789, "grad_norm": 0.73828125, "learning_rate": 0.0003095, "loss": 5.9906, "mean_token_accuracy": 0.1714184045791626, "num_tokens": 10082129.0, "step": 620 }, { "entropy": 6.200570106506348, "epoch": 0.5112474437627812, "grad_norm": 0.7109375, "learning_rate": 0.000312, "loss": 5.9749, "mean_token_accuracy": 0.1746130794286728, "num_tokens": 10163800.0, "step": 625 }, { "entropy": 6.2643028736114506, "epoch": 0.5153374233128835, "grad_norm": 0.79296875, "learning_rate": 0.0003145, "loss": 5.987, "mean_token_accuracy": 0.17392124682664872, "num_tokens": 10245046.0, "step": 630 }, { "entropy": 6.170021629333496, "epoch": 0.5194274028629857, "grad_norm": 0.7890625, "learning_rate": 0.000317, "loss": 5.9455, "mean_token_accuracy": 0.17663631588220596, "num_tokens": 10326555.0, "step": 635 }, { "entropy": 6.219466257095337, "epoch": 0.523517382413088, "grad_norm": 0.72265625, "learning_rate": 0.0003195, "loss": 5.9848, "mean_token_accuracy": 0.17208182513713838, "num_tokens": 10407844.0, "step": 640 }, { "entropy": 6.197711086273193, "epoch": 0.5276073619631901, "grad_norm": 0.7734375, "learning_rate": 0.000322, "loss": 5.9154, "mean_token_accuracy": 0.17427301406860352, "num_tokens": 10489166.0, "step": 645 }, { "entropy": 6.149255561828613, "epoch": 0.5316973415132924, "grad_norm": 0.765625, "learning_rate": 0.00032450000000000003, "loss": 5.9272, "mean_token_accuracy": 0.17766241431236268, "num_tokens": 10570741.0, "step": 650 }, { "entropy": 6.157588863372803, "epoch": 0.5357873210633947, "grad_norm": 0.74609375, "learning_rate": 0.00032700000000000003, "loss": 5.8674, "mean_token_accuracy": 0.17878217101097107, "num_tokens": 10651421.0, "step": 655 }, { "entropy": 6.161151552200318, "epoch": 0.5398773006134969, "grad_norm": 0.73828125, "learning_rate": 0.00032950000000000004, "loss": 5.938, "mean_token_accuracy": 0.17769655585289001, "num_tokens": 10732901.0, "step": 660 }, { "entropy": 6.172371435165405, "epoch": 0.5439672801635992, "grad_norm": 0.69921875, "learning_rate": 0.00033200000000000005, "loss": 5.9471, "mean_token_accuracy": 0.17350029200315475, "num_tokens": 10814362.0, "step": 665 }, { "entropy": 6.131135034561157, "epoch": 0.5480572597137015, "grad_norm": 0.72265625, "learning_rate": 0.00033450000000000005, "loss": 5.8894, "mean_token_accuracy": 0.18017593771219254, "num_tokens": 10896021.0, "step": 670 }, { "entropy": 6.120968770980835, "epoch": 0.5521472392638037, "grad_norm": 0.765625, "learning_rate": 0.000337, "loss": 5.8905, "mean_token_accuracy": 0.17684736251831054, "num_tokens": 10977596.0, "step": 675 }, { "entropy": 6.081154823303223, "epoch": 0.556237218813906, "grad_norm": 0.78125, "learning_rate": 0.0003395, "loss": 5.8466, "mean_token_accuracy": 0.17903628051280976, "num_tokens": 11058660.0, "step": 680 }, { "entropy": 6.145718908309936, "epoch": 0.5603271983640081, "grad_norm": 0.69140625, "learning_rate": 0.000342, "loss": 5.9027, "mean_token_accuracy": 0.17930294275283815, "num_tokens": 11140110.0, "step": 685 }, { "entropy": 6.064173078536987, "epoch": 0.5644171779141104, "grad_norm": 0.765625, "learning_rate": 0.00034449999999999997, "loss": 5.8366, "mean_token_accuracy": 0.18158787190914155, "num_tokens": 11221772.0, "step": 690 }, { "entropy": 6.099767875671387, "epoch": 0.5685071574642127, "grad_norm": 0.78125, "learning_rate": 0.000347, "loss": 5.8539, "mean_token_accuracy": 0.17807960510253906, "num_tokens": 11303047.0, "step": 695 }, { "entropy": 6.058339786529541, "epoch": 0.5725971370143149, "grad_norm": 0.71875, "learning_rate": 0.0003495, "loss": 5.8554, "mean_token_accuracy": 0.18217273205518722, "num_tokens": 11384146.0, "step": 700 }, { "entropy": 6.080018758773804, "epoch": 0.5766871165644172, "grad_norm": 0.89453125, "learning_rate": 0.000352, "loss": 5.8803, "mean_token_accuracy": 0.1770617514848709, "num_tokens": 11465561.0, "step": 705 }, { "entropy": 6.095365285873413, "epoch": 0.5807770961145194, "grad_norm": 0.8359375, "learning_rate": 0.0003545, "loss": 5.8922, "mean_token_accuracy": 0.17715113013982772, "num_tokens": 11547242.0, "step": 710 }, { "entropy": 6.099693870544433, "epoch": 0.5848670756646217, "grad_norm": 0.703125, "learning_rate": 0.000357, "loss": 5.8803, "mean_token_accuracy": 0.17778826057910918, "num_tokens": 11628796.0, "step": 715 }, { "entropy": 6.091752576828003, "epoch": 0.588957055214724, "grad_norm": 0.69140625, "learning_rate": 0.0003595, "loss": 5.8647, "mean_token_accuracy": 0.177607761323452, "num_tokens": 11709934.0, "step": 720 }, { "entropy": 6.030358505249024, "epoch": 0.5930470347648262, "grad_norm": 0.69140625, "learning_rate": 0.000362, "loss": 5.8122, "mean_token_accuracy": 0.1837711438536644, "num_tokens": 11791434.0, "step": 725 }, { "entropy": 6.027116060256958, "epoch": 0.5971370143149284, "grad_norm": 0.65234375, "learning_rate": 0.0003645, "loss": 5.7952, "mean_token_accuracy": 0.18036992698907853, "num_tokens": 11873110.0, "step": 730 }, { "entropy": 6.058136081695556, "epoch": 0.6012269938650306, "grad_norm": 0.6953125, "learning_rate": 0.000367, "loss": 5.8507, "mean_token_accuracy": 0.1789834216237068, "num_tokens": 11953641.0, "step": 735 }, { "entropy": 6.024233865737915, "epoch": 0.6053169734151329, "grad_norm": 0.734375, "learning_rate": 0.0003695, "loss": 5.8079, "mean_token_accuracy": 0.1811279147863388, "num_tokens": 12034832.0, "step": 740 }, { "entropy": 6.063544702529907, "epoch": 0.6094069529652352, "grad_norm": 0.78515625, "learning_rate": 0.000372, "loss": 5.8226, "mean_token_accuracy": 0.17915984243154526, "num_tokens": 12116448.0, "step": 745 }, { "entropy": 6.08197922706604, "epoch": 0.6134969325153374, "grad_norm": 0.72265625, "learning_rate": 0.0003745, "loss": 5.8792, "mean_token_accuracy": 0.17911573946475984, "num_tokens": 12197613.0, "step": 750 }, { "entropy": 5.976650238037109, "epoch": 0.6175869120654397, "grad_norm": 0.71875, "learning_rate": 0.000377, "loss": 5.7948, "mean_token_accuracy": 0.18249804377555848, "num_tokens": 12278678.0, "step": 755 }, { "entropy": 6.024249076843262, "epoch": 0.621676891615542, "grad_norm": 0.73828125, "learning_rate": 0.0003795, "loss": 5.7897, "mean_token_accuracy": 0.18367442935705186, "num_tokens": 12359872.0, "step": 760 }, { "entropy": 6.026234292984009, "epoch": 0.6257668711656442, "grad_norm": 0.65234375, "learning_rate": 0.000382, "loss": 5.8135, "mean_token_accuracy": 0.17825709283351898, "num_tokens": 12441096.0, "step": 765 }, { "entropy": 5.9644246101379395, "epoch": 0.6298568507157464, "grad_norm": 0.71875, "learning_rate": 0.0003845, "loss": 5.7459, "mean_token_accuracy": 0.1821851447224617, "num_tokens": 12522312.0, "step": 770 }, { "entropy": 5.9929276466369625, "epoch": 0.6339468302658486, "grad_norm": 0.6796875, "learning_rate": 0.00038700000000000003, "loss": 5.7928, "mean_token_accuracy": 0.18280912786722184, "num_tokens": 12603618.0, "step": 775 }, { "entropy": 5.982466363906861, "epoch": 0.6380368098159509, "grad_norm": 0.796875, "learning_rate": 0.00038950000000000003, "loss": 5.7892, "mean_token_accuracy": 0.18169402629137038, "num_tokens": 12685152.0, "step": 780 }, { "entropy": 6.002019023895263, "epoch": 0.6421267893660532, "grad_norm": 0.99609375, "learning_rate": 0.00039200000000000004, "loss": 5.7585, "mean_token_accuracy": 0.1856043517589569, "num_tokens": 12766846.0, "step": 785 }, { "entropy": 5.955765247344971, "epoch": 0.6462167689161554, "grad_norm": 0.69140625, "learning_rate": 0.00039450000000000005, "loss": 5.7334, "mean_token_accuracy": 0.18403113782405853, "num_tokens": 12848427.0, "step": 790 }, { "entropy": 5.9625551223754885, "epoch": 0.6503067484662577, "grad_norm": 0.7578125, "learning_rate": 0.00039700000000000005, "loss": 5.7819, "mean_token_accuracy": 0.18669268637895584, "num_tokens": 12929737.0, "step": 795 }, { "entropy": 5.9695086002349855, "epoch": 0.65439672801636, "grad_norm": 0.6484375, "learning_rate": 0.0003995, "loss": 5.7912, "mean_token_accuracy": 0.1820902943611145, "num_tokens": 13011108.0, "step": 800 }, { "entropy": 6.0248494148254395, "epoch": 0.6584867075664622, "grad_norm": 0.70703125, "learning_rate": 0.000402, "loss": 5.8197, "mean_token_accuracy": 0.17902762889862062, "num_tokens": 13092403.0, "step": 805 }, { "entropy": 5.980278158187867, "epoch": 0.6625766871165644, "grad_norm": 0.7109375, "learning_rate": 0.0004045, "loss": 5.7867, "mean_token_accuracy": 0.179145847260952, "num_tokens": 13174002.0, "step": 810 }, { "entropy": 5.95001859664917, "epoch": 0.6666666666666666, "grad_norm": 0.7578125, "learning_rate": 0.00040699999999999997, "loss": 5.7316, "mean_token_accuracy": 0.18538380563259124, "num_tokens": 13255596.0, "step": 815 }, { "entropy": 5.877342081069946, "epoch": 0.6707566462167689, "grad_norm": 0.72265625, "learning_rate": 0.0004095, "loss": 5.7142, "mean_token_accuracy": 0.18710004389286042, "num_tokens": 13337275.0, "step": 820 }, { "entropy": 5.950238132476807, "epoch": 0.6748466257668712, "grad_norm": 0.71484375, "learning_rate": 0.000412, "loss": 5.7579, "mean_token_accuracy": 0.18248329907655716, "num_tokens": 13418839.0, "step": 825 }, { "entropy": 5.902752208709717, "epoch": 0.6789366053169734, "grad_norm": 0.6328125, "learning_rate": 0.0004145, "loss": 5.7014, "mean_token_accuracy": 0.1873370885848999, "num_tokens": 13500506.0, "step": 830 }, { "entropy": 5.974310207366943, "epoch": 0.6830265848670757, "grad_norm": 0.8125, "learning_rate": 0.000417, "loss": 5.7547, "mean_token_accuracy": 0.18390790075063707, "num_tokens": 13581825.0, "step": 835 }, { "entropy": 5.879995679855346, "epoch": 0.6871165644171779, "grad_norm": 0.68359375, "learning_rate": 0.0004195, "loss": 5.6878, "mean_token_accuracy": 0.18741529285907746, "num_tokens": 13662792.0, "step": 840 }, { "entropy": 5.954990816116333, "epoch": 0.6912065439672802, "grad_norm": 0.70703125, "learning_rate": 0.000422, "loss": 5.7638, "mean_token_accuracy": 0.18056032210588455, "num_tokens": 13744054.0, "step": 845 }, { "entropy": 5.853623104095459, "epoch": 0.6952965235173824, "grad_norm": 0.7421875, "learning_rate": 0.0004245, "loss": 5.6782, "mean_token_accuracy": 0.18856651484966278, "num_tokens": 13825657.0, "step": 850 }, { "entropy": 5.890990686416626, "epoch": 0.6993865030674846, "grad_norm": 0.69921875, "learning_rate": 0.000427, "loss": 5.7072, "mean_token_accuracy": 0.18802714049816133, "num_tokens": 13907176.0, "step": 855 }, { "entropy": 5.901055812835693, "epoch": 0.7034764826175869, "grad_norm": 0.69140625, "learning_rate": 0.0004295, "loss": 5.7033, "mean_token_accuracy": 0.18591031730175017, "num_tokens": 13988341.0, "step": 860 }, { "entropy": 5.88054780960083, "epoch": 0.7075664621676891, "grad_norm": 0.69921875, "learning_rate": 0.000432, "loss": 5.7229, "mean_token_accuracy": 0.1866290345788002, "num_tokens": 14069996.0, "step": 865 }, { "entropy": 5.870758438110352, "epoch": 0.7116564417177914, "grad_norm": 0.75390625, "learning_rate": 0.0004345, "loss": 5.6977, "mean_token_accuracy": 0.18629909306764603, "num_tokens": 14151377.0, "step": 870 }, { "entropy": 5.87124457359314, "epoch": 0.7157464212678937, "grad_norm": 0.79296875, "learning_rate": 0.000437, "loss": 5.6897, "mean_token_accuracy": 0.18838011622428893, "num_tokens": 14232636.0, "step": 875 }, { "entropy": 5.844027185440064, "epoch": 0.7198364008179959, "grad_norm": 0.671875, "learning_rate": 0.0004395, "loss": 5.6779, "mean_token_accuracy": 0.18566754907369615, "num_tokens": 14314309.0, "step": 880 }, { "entropy": 5.842675304412841, "epoch": 0.7239263803680982, "grad_norm": 0.671875, "learning_rate": 0.000442, "loss": 5.6504, "mean_token_accuracy": 0.189619480073452, "num_tokens": 14395936.0, "step": 885 }, { "entropy": 5.913097095489502, "epoch": 0.7280163599182005, "grad_norm": 0.6796875, "learning_rate": 0.0004445, "loss": 5.6854, "mean_token_accuracy": 0.18613518625497819, "num_tokens": 14477308.0, "step": 890 }, { "entropy": 5.816605186462402, "epoch": 0.7321063394683026, "grad_norm": 0.67578125, "learning_rate": 0.000447, "loss": 5.6487, "mean_token_accuracy": 0.19115502685308455, "num_tokens": 14558960.0, "step": 895 }, { "entropy": 5.8206874370574955, "epoch": 0.7361963190184049, "grad_norm": 0.671875, "learning_rate": 0.00044950000000000003, "loss": 5.6503, "mean_token_accuracy": 0.18675681203603745, "num_tokens": 14638937.0, "step": 900 }, { "entropy": 5.848356771469116, "epoch": 0.7402862985685071, "grad_norm": 0.69921875, "learning_rate": 0.00045200000000000004, "loss": 5.6673, "mean_token_accuracy": 0.18895538300275802, "num_tokens": 14719760.0, "step": 905 }, { "entropy": 5.797045040130615, "epoch": 0.7443762781186094, "grad_norm": 0.71875, "learning_rate": 0.00045450000000000004, "loss": 5.6212, "mean_token_accuracy": 0.19115596413612365, "num_tokens": 14801266.0, "step": 910 }, { "entropy": 5.849314260482788, "epoch": 0.7484662576687117, "grad_norm": 0.67578125, "learning_rate": 0.00045700000000000005, "loss": 5.6874, "mean_token_accuracy": 0.18890151828527452, "num_tokens": 14882766.0, "step": 915 }, { "entropy": 5.789092397689819, "epoch": 0.7525562372188139, "grad_norm": 0.6953125, "learning_rate": 0.00045950000000000006, "loss": 5.6504, "mean_token_accuracy": 0.18801938593387604, "num_tokens": 14964343.0, "step": 920 }, { "entropy": 5.800046920776367, "epoch": 0.7566462167689162, "grad_norm": 0.65625, "learning_rate": 0.000462, "loss": 5.6093, "mean_token_accuracy": 0.1909247636795044, "num_tokens": 15045937.0, "step": 925 }, { "entropy": 5.792387008666992, "epoch": 0.7607361963190185, "grad_norm": 0.6328125, "learning_rate": 0.0004645, "loss": 5.6218, "mean_token_accuracy": 0.19211873561143875, "num_tokens": 15126836.0, "step": 930 }, { "entropy": 5.822949647903442, "epoch": 0.7648261758691206, "grad_norm": 0.67578125, "learning_rate": 0.000467, "loss": 5.6385, "mean_token_accuracy": 0.19017553627490996, "num_tokens": 15208384.0, "step": 935 }, { "entropy": 5.772562408447266, "epoch": 0.7689161554192229, "grad_norm": 0.69140625, "learning_rate": 0.0004695, "loss": 5.6223, "mean_token_accuracy": 0.18999192416667937, "num_tokens": 15288828.0, "step": 940 }, { "entropy": 5.786653804779053, "epoch": 0.7730061349693251, "grad_norm": 0.71484375, "learning_rate": 0.000472, "loss": 5.6312, "mean_token_accuracy": 0.1895082637667656, "num_tokens": 15370454.0, "step": 945 }, { "entropy": 5.792725038528443, "epoch": 0.7770961145194274, "grad_norm": 0.66015625, "learning_rate": 0.0004745, "loss": 5.6261, "mean_token_accuracy": 0.18655621856451035, "num_tokens": 15452077.0, "step": 950 }, { "entropy": 5.79119381904602, "epoch": 0.7811860940695297, "grad_norm": 0.69140625, "learning_rate": 0.000477, "loss": 5.6403, "mean_token_accuracy": 0.19076757580041886, "num_tokens": 15533553.0, "step": 955 }, { "entropy": 5.721160888671875, "epoch": 0.7852760736196319, "grad_norm": 0.68359375, "learning_rate": 0.0004795, "loss": 5.576, "mean_token_accuracy": 0.1896376132965088, "num_tokens": 15614710.0, "step": 960 }, { "entropy": 5.763787698745728, "epoch": 0.7893660531697342, "grad_norm": 0.6875, "learning_rate": 0.000482, "loss": 5.6253, "mean_token_accuracy": 0.19005389511585236, "num_tokens": 15696287.0, "step": 965 }, { "entropy": 5.74923620223999, "epoch": 0.7934560327198364, "grad_norm": 0.68359375, "learning_rate": 0.0004845, "loss": 5.5765, "mean_token_accuracy": 0.19157841056585312, "num_tokens": 15777635.0, "step": 970 }, { "entropy": 5.765731239318848, "epoch": 0.7975460122699386, "grad_norm": 0.625, "learning_rate": 0.000487, "loss": 5.6059, "mean_token_accuracy": 0.18815757483243942, "num_tokens": 15859321.0, "step": 975 }, { "entropy": 5.8039552688598635, "epoch": 0.8016359918200409, "grad_norm": 0.63671875, "learning_rate": 0.0004895, "loss": 5.6528, "mean_token_accuracy": 0.18570050597190857, "num_tokens": 15939821.0, "step": 980 }, { "entropy": 5.7897364616394045, "epoch": 0.8057259713701431, "grad_norm": 0.70703125, "learning_rate": 0.000492, "loss": 5.6536, "mean_token_accuracy": 0.18991607278585435, "num_tokens": 16020937.0, "step": 985 }, { "entropy": 5.74057035446167, "epoch": 0.8098159509202454, "grad_norm": 0.6328125, "learning_rate": 0.0004945, "loss": 5.5987, "mean_token_accuracy": 0.19221560955047606, "num_tokens": 16101964.0, "step": 990 }, { "entropy": 5.738554191589356, "epoch": 0.8139059304703476, "grad_norm": 0.73046875, "learning_rate": 0.000497, "loss": 5.571, "mean_token_accuracy": 0.1946096107363701, "num_tokens": 16183239.0, "step": 995 }, { "entropy": 5.774260759353638, "epoch": 0.8179959100204499, "grad_norm": 0.6796875, "learning_rate": 0.0004995, "loss": 5.619, "mean_token_accuracy": 0.18836521208286286, "num_tokens": 16264916.0, "step": 1000 }, { "epoch": 0.8179959100204499, "eval_entropy": 5.620751747293362, "eval_loss": 5.542455196380615, "eval_mean_token_accuracy": 0.1962628694804939, "eval_num_tokens": 16264916.0, "eval_runtime": 66.7485, "eval_samples_per_second": 61.979, "eval_steps_per_second": 7.76, "step": 1000 }, { "entropy": 5.703431701660156, "epoch": 0.8220858895705522, "grad_norm": 0.62890625, "learning_rate": 0.0004999998588806748, "loss": 5.5729, "mean_token_accuracy": 0.19113412499427795, "num_tokens": 16346636.0, "step": 1005 }, { "entropy": 5.726306343078614, "epoch": 0.8261758691206544, "grad_norm": 0.625, "learning_rate": 0.0004999992855837193, "loss": 5.596, "mean_token_accuracy": 0.19352659285068513, "num_tokens": 16428277.0, "step": 1010 }, { "entropy": 5.803088760375976, "epoch": 0.8302658486707567, "grad_norm": 0.66796875, "learning_rate": 0.0004999982712902986, "loss": 5.6477, "mean_token_accuracy": 0.18696338087320327, "num_tokens": 16509217.0, "step": 1015 }, { "entropy": 5.747509336471557, "epoch": 0.8343558282208589, "grad_norm": 0.6953125, "learning_rate": 0.0004999968160024007, "loss": 5.5741, "mean_token_accuracy": 0.1908252865076065, "num_tokens": 16590124.0, "step": 1020 }, { "entropy": 5.731025695800781, "epoch": 0.8384458077709611, "grad_norm": 0.6484375, "learning_rate": 0.000499994919722878, "loss": 5.5917, "mean_token_accuracy": 0.1903747096657753, "num_tokens": 16671723.0, "step": 1025 }, { "entropy": 5.742538118362427, "epoch": 0.8425357873210634, "grad_norm": 0.64453125, "learning_rate": 0.000499992582455447, "loss": 5.57, "mean_token_accuracy": 0.19232080429792403, "num_tokens": 16752874.0, "step": 1030 }, { "entropy": 5.69538893699646, "epoch": 0.8466257668711656, "grad_norm": 0.62890625, "learning_rate": 0.0004999898042046889, "loss": 5.5504, "mean_token_accuracy": 0.19378954619169236, "num_tokens": 16834468.0, "step": 1035 }, { "entropy": 5.696932601928711, "epoch": 0.8507157464212679, "grad_norm": 0.6796875, "learning_rate": 0.0004999865849760489, "loss": 5.5659, "mean_token_accuracy": 0.19112324118614196, "num_tokens": 16915209.0, "step": 1040 }, { "entropy": 5.734558725357056, "epoch": 0.8548057259713702, "grad_norm": 0.66015625, "learning_rate": 0.0004999829247758369, "loss": 5.5901, "mean_token_accuracy": 0.19213886857032775, "num_tokens": 16996383.0, "step": 1045 }, { "entropy": 5.696015024185181, "epoch": 0.8588957055214724, "grad_norm": 0.640625, "learning_rate": 0.0004999788236112266, "loss": 5.5468, "mean_token_accuracy": 0.19250663965940476, "num_tokens": 17077977.0, "step": 1050 }, { "entropy": 5.674211740493774, "epoch": 0.8629856850715747, "grad_norm": 0.68359375, "learning_rate": 0.0004999742814902564, "loss": 5.5486, "mean_token_accuracy": 0.1906023606657982, "num_tokens": 17159598.0, "step": 1055 }, { "entropy": 5.687230396270752, "epoch": 0.8670756646216768, "grad_norm": 0.61328125, "learning_rate": 0.0004999692984218287, "loss": 5.5454, "mean_token_accuracy": 0.19375290125608444, "num_tokens": 17241186.0, "step": 1060 }, { "entropy": 5.680391120910644, "epoch": 0.8711656441717791, "grad_norm": 0.63671875, "learning_rate": 0.0004999638744157103, "loss": 5.5283, "mean_token_accuracy": 0.19310349822044373, "num_tokens": 17322359.0, "step": 1065 }, { "entropy": 5.648583698272705, "epoch": 0.8752556237218814, "grad_norm": 0.64453125, "learning_rate": 0.0004999580094825323, "loss": 5.5011, "mean_token_accuracy": 0.1941932126879692, "num_tokens": 17403666.0, "step": 1070 }, { "entropy": 5.633043670654297, "epoch": 0.8793456032719836, "grad_norm": 0.63671875, "learning_rate": 0.0004999517036337897, "loss": 5.4866, "mean_token_accuracy": 0.1960804805159569, "num_tokens": 17484598.0, "step": 1075 }, { "entropy": 5.723490905761719, "epoch": 0.8834355828220859, "grad_norm": 0.67578125, "learning_rate": 0.0004999449568818422, "loss": 5.5704, "mean_token_accuracy": 0.19220139533281327, "num_tokens": 17566007.0, "step": 1080 }, { "entropy": 5.708897304534912, "epoch": 0.8875255623721882, "grad_norm": 0.60546875, "learning_rate": 0.0004999377692399132, "loss": 5.5831, "mean_token_accuracy": 0.19491149932146073, "num_tokens": 17647579.0, "step": 1085 }, { "entropy": 5.689042949676514, "epoch": 0.8916155419222904, "grad_norm": 0.62890625, "learning_rate": 0.0004999301407220903, "loss": 5.5509, "mean_token_accuracy": 0.19352951049804687, "num_tokens": 17728737.0, "step": 1090 }, { "entropy": 5.67222785949707, "epoch": 0.8957055214723927, "grad_norm": 0.62890625, "learning_rate": 0.0004999220713433255, "loss": 5.5014, "mean_token_accuracy": 0.19424259215593337, "num_tokens": 17809840.0, "step": 1095 }, { "entropy": 5.652796411514283, "epoch": 0.8997955010224948, "grad_norm": 0.640625, "learning_rate": 0.0004999135611194346, "loss": 5.5261, "mean_token_accuracy": 0.19446621388196944, "num_tokens": 17890911.0, "step": 1100 }, { "entropy": 5.596818733215332, "epoch": 0.9038854805725971, "grad_norm": 0.625, "learning_rate": 0.0004999046100670977, "loss": 5.4442, "mean_token_accuracy": 0.19822182655334472, "num_tokens": 17972009.0, "step": 1105 }, { "entropy": 5.6312689781188965, "epoch": 0.9079754601226994, "grad_norm": 0.671875, "learning_rate": 0.0004998952182038587, "loss": 5.5152, "mean_token_accuracy": 0.19492827951908112, "num_tokens": 18053732.0, "step": 1110 }, { "entropy": 5.628683233261109, "epoch": 0.9120654396728016, "grad_norm": 0.64453125, "learning_rate": 0.0004998853855481255, "loss": 5.4875, "mean_token_accuracy": 0.19863843619823457, "num_tokens": 18135254.0, "step": 1115 }, { "entropy": 5.69710865020752, "epoch": 0.9161554192229039, "grad_norm": 0.63671875, "learning_rate": 0.0004998751121191701, "loss": 5.5757, "mean_token_accuracy": 0.19335949271917344, "num_tokens": 18215113.0, "step": 1120 }, { "entropy": 5.6402688980102536, "epoch": 0.9202453987730062, "grad_norm": 0.6171875, "learning_rate": 0.0004998643979371283, "loss": 5.508, "mean_token_accuracy": 0.19808033257722854, "num_tokens": 18296389.0, "step": 1125 }, { "entropy": 5.586647415161133, "epoch": 0.9243353783231084, "grad_norm": 0.67578125, "learning_rate": 0.0004998532430229997, "loss": 5.4521, "mean_token_accuracy": 0.19976601600646973, "num_tokens": 18377575.0, "step": 1130 }, { "entropy": 5.6341845989227295, "epoch": 0.9284253578732107, "grad_norm": 0.5625, "learning_rate": 0.0004998416473986478, "loss": 5.5092, "mean_token_accuracy": 0.19597392678260803, "num_tokens": 18458950.0, "step": 1135 }, { "entropy": 5.594471788406372, "epoch": 0.9325153374233128, "grad_norm": 0.60546875, "learning_rate": 0.0004998296110868002, "loss": 5.4211, "mean_token_accuracy": 0.2008963093161583, "num_tokens": 18540068.0, "step": 1140 }, { "entropy": 5.568839168548584, "epoch": 0.9366053169734151, "grad_norm": 0.59765625, "learning_rate": 0.0004998171341110477, "loss": 5.449, "mean_token_accuracy": 0.2023203492164612, "num_tokens": 18621708.0, "step": 1145 }, { "entropy": 5.673064708709717, "epoch": 0.9406952965235174, "grad_norm": 0.578125, "learning_rate": 0.0004998042164958452, "loss": 5.5271, "mean_token_accuracy": 0.19440514743328094, "num_tokens": 18702656.0, "step": 1150 }, { "entropy": 5.566248273849487, "epoch": 0.9447852760736196, "grad_norm": 0.5703125, "learning_rate": 0.000499790858266511, "loss": 5.4517, "mean_token_accuracy": 0.19680088460445405, "num_tokens": 18784248.0, "step": 1155 }, { "entropy": 5.588905477523804, "epoch": 0.9488752556237219, "grad_norm": 0.609375, "learning_rate": 0.0004997770594492271, "loss": 5.4341, "mean_token_accuracy": 0.20149867236614227, "num_tokens": 18865347.0, "step": 1160 }, { "entropy": 5.639943552017212, "epoch": 0.9529652351738241, "grad_norm": 0.578125, "learning_rate": 0.0004997628200710391, "loss": 5.4756, "mean_token_accuracy": 0.19862927943468095, "num_tokens": 18946875.0, "step": 1165 }, { "entropy": 5.597586345672608, "epoch": 0.9570552147239264, "grad_norm": 0.59375, "learning_rate": 0.0004997481401598561, "loss": 5.442, "mean_token_accuracy": 0.19850796461105347, "num_tokens": 19027484.0, "step": 1170 }, { "entropy": 5.579336309432984, "epoch": 0.9611451942740287, "grad_norm": 0.55859375, "learning_rate": 0.0004997330197444505, "loss": 5.4454, "mean_token_accuracy": 0.19992446452379226, "num_tokens": 19109116.0, "step": 1175 }, { "entropy": 5.627445363998413, "epoch": 0.9652351738241309, "grad_norm": 0.59375, "learning_rate": 0.0004997174588544581, "loss": 5.4881, "mean_token_accuracy": 0.1974893182516098, "num_tokens": 19190124.0, "step": 1180 }, { "entropy": 5.580318975448608, "epoch": 0.9693251533742331, "grad_norm": 0.6328125, "learning_rate": 0.0004997014575203781, "loss": 5.477, "mean_token_accuracy": 0.19553500413894653, "num_tokens": 19271429.0, "step": 1185 }, { "entropy": 5.659302473068237, "epoch": 0.9734151329243353, "grad_norm": 0.56640625, "learning_rate": 0.0004996850157735732, "loss": 5.5329, "mean_token_accuracy": 0.19281953871250151, "num_tokens": 19352734.0, "step": 1190 }, { "entropy": 5.575047349929809, "epoch": 0.9775051124744376, "grad_norm": 0.62109375, "learning_rate": 0.0004996681336462691, "loss": 5.4124, "mean_token_accuracy": 0.20342059433460236, "num_tokens": 19434350.0, "step": 1195 }, { "entropy": 5.59806718826294, "epoch": 0.9815950920245399, "grad_norm": 0.61328125, "learning_rate": 0.0004996508111715541, "loss": 5.4751, "mean_token_accuracy": 0.19599334448575972, "num_tokens": 19515490.0, "step": 1200 }, { "entropy": 5.565750408172607, "epoch": 0.9856850715746421, "grad_norm": 0.60546875, "learning_rate": 0.0004996330483833805, "loss": 5.4256, "mean_token_accuracy": 0.19978437274694444, "num_tokens": 19597108.0, "step": 1205 }, { "entropy": 5.583062028884887, "epoch": 0.9897750511247444, "grad_norm": 0.5703125, "learning_rate": 0.0004996148453165632, "loss": 5.4578, "mean_token_accuracy": 0.20042912364006044, "num_tokens": 19678476.0, "step": 1210 }, { "entropy": 5.574772119522095, "epoch": 0.9938650306748467, "grad_norm": 0.56640625, "learning_rate": 0.0004995962020067798, "loss": 5.4756, "mean_token_accuracy": 0.19866746962070464, "num_tokens": 19760179.0, "step": 1215 }, { "entropy": 5.588492012023925, "epoch": 0.9979550102249489, "grad_norm": 0.58203125, "learning_rate": 0.0004995771184905714, "loss": 5.4431, "mean_token_accuracy": 0.19795191138982773, "num_tokens": 19841815.0, "step": 1220 }, { "entropy": 5.549735175238715, "epoch": 1.0016359918200408, "grad_norm": 0.6171875, "learning_rate": 0.000499557594805341, "loss": 5.3939, "mean_token_accuracy": 0.20071937474939558, "num_tokens": 19907854.0, "step": 1225 }, { "entropy": 5.542763519287109, "epoch": 1.0057259713701432, "grad_norm": 0.58984375, "learning_rate": 0.0004995376309893553, "loss": 5.3708, "mean_token_accuracy": 0.2003950893878937, "num_tokens": 19988742.0, "step": 1230 }, { "entropy": 5.556063175201416, "epoch": 1.0098159509202453, "grad_norm": 0.55859375, "learning_rate": 0.0004995172270817429, "loss": 5.3388, "mean_token_accuracy": 0.20662298053503036, "num_tokens": 20070182.0, "step": 1235 }, { "entropy": 5.560736560821534, "epoch": 1.0139059304703477, "grad_norm": 0.60546875, "learning_rate": 0.0004994963831224954, "loss": 5.3558, "mean_token_accuracy": 0.20254284143447876, "num_tokens": 20151791.0, "step": 1240 }, { "entropy": 5.546072101593017, "epoch": 1.0179959100204499, "grad_norm": 0.59765625, "learning_rate": 0.0004994750991524667, "loss": 5.3976, "mean_token_accuracy": 0.1990876540541649, "num_tokens": 20233472.0, "step": 1245 }, { "entropy": 5.530217123031616, "epoch": 1.0220858895705522, "grad_norm": 0.60546875, "learning_rate": 0.0004994533752133732, "loss": 5.3601, "mean_token_accuracy": 0.20016493499279023, "num_tokens": 20314857.0, "step": 1250 }, { "entropy": 5.543700551986694, "epoch": 1.0261758691206544, "grad_norm": 0.62890625, "learning_rate": 0.0004994312113477936, "loss": 5.352, "mean_token_accuracy": 0.20525595247745515, "num_tokens": 20396285.0, "step": 1255 }, { "entropy": 5.514524841308594, "epoch": 1.0302658486707565, "grad_norm": 0.68359375, "learning_rate": 0.000499408607599169, "loss": 5.3481, "mean_token_accuracy": 0.20686980634927749, "num_tokens": 20477967.0, "step": 1260 }, { "entropy": 5.527777147293091, "epoch": 1.034355828220859, "grad_norm": 0.6796875, "learning_rate": 0.0004993855640118023, "loss": 5.3442, "mean_token_accuracy": 0.20495522618293763, "num_tokens": 20559529.0, "step": 1265 }, { "entropy": 5.49743971824646, "epoch": 1.038445807770961, "grad_norm": 0.58984375, "learning_rate": 0.0004993620806308589, "loss": 5.3518, "mean_token_accuracy": 0.2049548387527466, "num_tokens": 20641147.0, "step": 1270 }, { "entropy": 5.5427220344543455, "epoch": 1.0425357873210634, "grad_norm": 0.5703125, "learning_rate": 0.000499338157502366, "loss": 5.328, "mean_token_accuracy": 0.20616023540496825, "num_tokens": 20722330.0, "step": 1275 }, { "entropy": 5.5480578422546385, "epoch": 1.0466257668711656, "grad_norm": 0.6640625, "learning_rate": 0.0004993137946732125, "loss": 5.3861, "mean_token_accuracy": 0.20277144610881806, "num_tokens": 20803908.0, "step": 1280 }, { "entropy": 5.509058952331543, "epoch": 1.050715746421268, "grad_norm": 0.59375, "learning_rate": 0.0004992889921911494, "loss": 5.3373, "mean_token_accuracy": 0.20504923164844513, "num_tokens": 20885126.0, "step": 1285 }, { "entropy": 5.523485803604126, "epoch": 1.0548057259713701, "grad_norm": 0.60546875, "learning_rate": 0.0004992637501047895, "loss": 5.3481, "mean_token_accuracy": 0.2020747557282448, "num_tokens": 20966557.0, "step": 1290 }, { "entropy": 5.484934377670288, "epoch": 1.0588957055214725, "grad_norm": 0.56640625, "learning_rate": 0.0004992380684636069, "loss": 5.3221, "mean_token_accuracy": 0.20487241446971893, "num_tokens": 21047765.0, "step": 1295 }, { "entropy": 5.465098142623901, "epoch": 1.0629856850715746, "grad_norm": 0.5859375, "learning_rate": 0.0004992119473179374, "loss": 5.2736, "mean_token_accuracy": 0.20992589145898818, "num_tokens": 21128728.0, "step": 1300 }, { "entropy": 5.525388574600219, "epoch": 1.0670756646216768, "grad_norm": 0.65625, "learning_rate": 0.0004991853867189782, "loss": 5.3678, "mean_token_accuracy": 0.20483604222536086, "num_tokens": 21210008.0, "step": 1305 }, { "entropy": 5.444643878936768, "epoch": 1.0711656441717792, "grad_norm": 0.625, "learning_rate": 0.0004991583867187878, "loss": 5.304, "mean_token_accuracy": 0.20822575688362122, "num_tokens": 21291654.0, "step": 1310 }, { "entropy": 5.526231336593628, "epoch": 1.0752556237218813, "grad_norm": 0.61328125, "learning_rate": 0.000499130947370286, "loss": 5.3501, "mean_token_accuracy": 0.20411286503076553, "num_tokens": 21372995.0, "step": 1315 }, { "entropy": 5.496540975570679, "epoch": 1.0793456032719837, "grad_norm": 0.56640625, "learning_rate": 0.0004991030687272537, "loss": 5.3221, "mean_token_accuracy": 0.20735106319189073, "num_tokens": 21453949.0, "step": 1320 }, { "entropy": 5.458919334411621, "epoch": 1.0834355828220859, "grad_norm": 0.5703125, "learning_rate": 0.0004990747508443325, "loss": 5.2914, "mean_token_accuracy": 0.2070190951228142, "num_tokens": 21535155.0, "step": 1325 }, { "entropy": 5.507519292831421, "epoch": 1.0875255623721882, "grad_norm": 0.58984375, "learning_rate": 0.0004990459937770257, "loss": 5.3517, "mean_token_accuracy": 0.20332252085208893, "num_tokens": 21616392.0, "step": 1330 }, { "entropy": 5.458401536941528, "epoch": 1.0916155419222904, "grad_norm": 0.61328125, "learning_rate": 0.0004990167975816963, "loss": 5.3111, "mean_token_accuracy": 0.2075095370411873, "num_tokens": 21697618.0, "step": 1335 }, { "entropy": 5.520397424697876, "epoch": 1.0957055214723925, "grad_norm": 0.62890625, "learning_rate": 0.000498987162315569, "loss": 5.3616, "mean_token_accuracy": 0.205839866399765, "num_tokens": 21779313.0, "step": 1340 }, { "entropy": 5.455747270584107, "epoch": 1.099795501022495, "grad_norm": 0.5859375, "learning_rate": 0.0004989570880367284, "loss": 5.3211, "mean_token_accuracy": 0.20768722593784333, "num_tokens": 21860494.0, "step": 1345 }, { "entropy": 5.502938556671142, "epoch": 1.103885480572597, "grad_norm": 0.59765625, "learning_rate": 0.0004989265748041199, "loss": 5.3221, "mean_token_accuracy": 0.20195858627557756, "num_tokens": 21942155.0, "step": 1350 }, { "entropy": 5.455473470687866, "epoch": 1.1079754601226994, "grad_norm": 0.55078125, "learning_rate": 0.000498895622677549, "loss": 5.2769, "mean_token_accuracy": 0.20793799608945845, "num_tokens": 22023808.0, "step": 1355 }, { "entropy": 5.413159942626953, "epoch": 1.1120654396728016, "grad_norm": 0.60546875, "learning_rate": 0.0004988642317176818, "loss": 5.2629, "mean_token_accuracy": 0.20979762077331543, "num_tokens": 22105395.0, "step": 1360 }, { "entropy": 5.447505235671997, "epoch": 1.116155419222904, "grad_norm": 0.57421875, "learning_rate": 0.0004988324019860442, "loss": 5.3046, "mean_token_accuracy": 0.20722273737192154, "num_tokens": 22186557.0, "step": 1365 }, { "entropy": 5.4436736583709715, "epoch": 1.120245398773006, "grad_norm": 0.5859375, "learning_rate": 0.0004988001335450221, "loss": 5.291, "mean_token_accuracy": 0.2108110785484314, "num_tokens": 22267743.0, "step": 1370 }, { "entropy": 5.4880999565124515, "epoch": 1.1243353783231085, "grad_norm": 0.6171875, "learning_rate": 0.0004987674264578615, "loss": 5.3204, "mean_token_accuracy": 0.20962023735046387, "num_tokens": 22348936.0, "step": 1375 }, { "entropy": 5.4847039699554445, "epoch": 1.1284253578732106, "grad_norm": 0.60546875, "learning_rate": 0.0004987342807886678, "loss": 5.3254, "mean_token_accuracy": 0.20564790517091752, "num_tokens": 22430536.0, "step": 1380 }, { "entropy": 5.445675849914551, "epoch": 1.132515337423313, "grad_norm": 0.59375, "learning_rate": 0.0004987006966024062, "loss": 5.2953, "mean_token_accuracy": 0.20681943744421005, "num_tokens": 22512171.0, "step": 1385 }, { "entropy": 5.41473798751831, "epoch": 1.1366053169734152, "grad_norm": 0.609375, "learning_rate": 0.0004986666739649016, "loss": 5.275, "mean_token_accuracy": 0.207230344414711, "num_tokens": 22593014.0, "step": 1390 }, { "entropy": 5.447560405731201, "epoch": 1.1406952965235173, "grad_norm": 0.57421875, "learning_rate": 0.0004986322129428378, "loss": 5.3019, "mean_token_accuracy": 0.20656066685914992, "num_tokens": 22674227.0, "step": 1395 }, { "entropy": 5.410824537277222, "epoch": 1.1447852760736197, "grad_norm": 0.578125, "learning_rate": 0.0004985973136037582, "loss": 5.2531, "mean_token_accuracy": 0.21179211437702178, "num_tokens": 22755657.0, "step": 1400 }, { "entropy": 5.422730493545532, "epoch": 1.1488752556237218, "grad_norm": 0.58203125, "learning_rate": 0.0004985619760160651, "loss": 5.2647, "mean_token_accuracy": 0.21022907346487046, "num_tokens": 22837022.0, "step": 1405 }, { "entropy": 5.468639755249024, "epoch": 1.1529652351738242, "grad_norm": 0.55859375, "learning_rate": 0.0004985262002490201, "loss": 5.3027, "mean_token_accuracy": 0.20850888937711715, "num_tokens": 22918645.0, "step": 1410 }, { "entropy": 5.419819974899292, "epoch": 1.1570552147239264, "grad_norm": 0.57421875, "learning_rate": 0.0004984899863727429, "loss": 5.2586, "mean_token_accuracy": 0.20885598808526992, "num_tokens": 22999780.0, "step": 1415 }, { "entropy": 5.411123847961425, "epoch": 1.1611451942740287, "grad_norm": 0.5546875, "learning_rate": 0.0004984533344582127, "loss": 5.2714, "mean_token_accuracy": 0.20758982598781586, "num_tokens": 23080822.0, "step": 1420 }, { "entropy": 5.4339988231658936, "epoch": 1.1652351738241309, "grad_norm": 0.62109375, "learning_rate": 0.0004984162445772668, "loss": 5.2977, "mean_token_accuracy": 0.20715069770812988, "num_tokens": 23161974.0, "step": 1425 }, { "entropy": 5.434486532211304, "epoch": 1.169325153374233, "grad_norm": 0.5546875, "learning_rate": 0.0004983787168026011, "loss": 5.302, "mean_token_accuracy": 0.20727487355470658, "num_tokens": 23242609.0, "step": 1430 }, { "entropy": 5.449370861053467, "epoch": 1.1734151329243354, "grad_norm": 0.578125, "learning_rate": 0.0004983407512077695, "loss": 5.2799, "mean_token_accuracy": 0.20913066565990449, "num_tokens": 23324079.0, "step": 1435 }, { "entropy": 5.4475157260894775, "epoch": 1.1775051124744376, "grad_norm": 0.55078125, "learning_rate": 0.0004983023478671843, "loss": 5.306, "mean_token_accuracy": 0.2085559546947479, "num_tokens": 23405378.0, "step": 1440 }, { "entropy": 5.435064792633057, "epoch": 1.18159509202454, "grad_norm": 0.5859375, "learning_rate": 0.0004982635068561159, "loss": 5.2771, "mean_token_accuracy": 0.20945741534233092, "num_tokens": 23486427.0, "step": 1445 }, { "entropy": 5.477257966995239, "epoch": 1.185685071574642, "grad_norm": 0.58203125, "learning_rate": 0.0004982242282506918, "loss": 5.3183, "mean_token_accuracy": 0.20545368492603303, "num_tokens": 23567636.0, "step": 1450 }, { "entropy": 5.36335973739624, "epoch": 1.1897750511247445, "grad_norm": 0.57421875, "learning_rate": 0.0004981845121278982, "loss": 5.2369, "mean_token_accuracy": 0.2115119993686676, "num_tokens": 23649150.0, "step": 1455 }, { "entropy": 5.4084007263183596, "epoch": 1.1938650306748466, "grad_norm": 0.64453125, "learning_rate": 0.0004981443585655781, "loss": 5.2562, "mean_token_accuracy": 0.21013395488262177, "num_tokens": 23730790.0, "step": 1460 }, { "entropy": 5.416132402420044, "epoch": 1.1979550102249488, "grad_norm": 0.625, "learning_rate": 0.0004981037676424321, "loss": 5.2793, "mean_token_accuracy": 0.20953387916088104, "num_tokens": 23812424.0, "step": 1465 }, { "entropy": 5.394218444824219, "epoch": 1.2020449897750511, "grad_norm": 0.7265625, "learning_rate": 0.0004980627394380182, "loss": 5.2301, "mean_token_accuracy": 0.20996304601430893, "num_tokens": 23894028.0, "step": 1470 }, { "entropy": 5.4422766208648685, "epoch": 1.2061349693251533, "grad_norm": 0.578125, "learning_rate": 0.0004980212740327509, "loss": 5.3159, "mean_token_accuracy": 0.2050097852945328, "num_tokens": 23974861.0, "step": 1475 }, { "entropy": 5.430482530593872, "epoch": 1.2102249488752557, "grad_norm": 0.6015625, "learning_rate": 0.0004979793715079024, "loss": 5.2795, "mean_token_accuracy": 0.20633518546819687, "num_tokens": 24055895.0, "step": 1480 }, { "entropy": 5.406202507019043, "epoch": 1.2143149284253578, "grad_norm": 0.5625, "learning_rate": 0.0004979370319456011, "loss": 5.2753, "mean_token_accuracy": 0.20794048756361008, "num_tokens": 24137482.0, "step": 1485 }, { "entropy": 5.3932568550109865, "epoch": 1.2184049079754602, "grad_norm": 0.5546875, "learning_rate": 0.0004978942554288321, "loss": 5.2644, "mean_token_accuracy": 0.21049268096685408, "num_tokens": 24218550.0, "step": 1490 }, { "entropy": 5.417111778259278, "epoch": 1.2224948875255623, "grad_norm": 0.57421875, "learning_rate": 0.000497851042041437, "loss": 5.2936, "mean_token_accuracy": 0.2090729996562004, "num_tokens": 24299695.0, "step": 1495 }, { "entropy": 5.4002760887146, "epoch": 1.2265848670756645, "grad_norm": 0.5625, "learning_rate": 0.0004978073918681135, "loss": 5.2376, "mean_token_accuracy": 0.21040473878383636, "num_tokens": 24380586.0, "step": 1500 }, { "epoch": 1.2265848670756645, "eval_entropy": 5.243330517330685, "eval_loss": 5.261579990386963, "eval_mean_token_accuracy": 0.21201344731921856, "eval_num_tokens": 24380586.0, "eval_runtime": 66.2105, "eval_samples_per_second": 62.483, "eval_steps_per_second": 7.824, "step": 1500 }, { "entropy": 5.380454349517822, "epoch": 1.2306748466257669, "grad_norm": 0.57421875, "learning_rate": 0.0004977633049944159, "loss": 5.2593, "mean_token_accuracy": 0.21092124432325363, "num_tokens": 24462173.0, "step": 1505 }, { "entropy": 5.418602037429809, "epoch": 1.234764826175869, "grad_norm": 0.56640625, "learning_rate": 0.0004977187815067538, "loss": 5.277, "mean_token_accuracy": 0.21084222346544265, "num_tokens": 24543390.0, "step": 1510 }, { "entropy": 5.48295521736145, "epoch": 1.2388548057259714, "grad_norm": 0.58984375, "learning_rate": 0.0004976738214923927, "loss": 5.3509, "mean_token_accuracy": 0.2032844677567482, "num_tokens": 24624474.0, "step": 1515 }, { "entropy": 5.410760164260864, "epoch": 1.2429447852760735, "grad_norm": 0.5859375, "learning_rate": 0.0004976284250394539, "loss": 5.2574, "mean_token_accuracy": 0.21011308580636978, "num_tokens": 24706059.0, "step": 1520 }, { "entropy": 5.403225946426391, "epoch": 1.247034764826176, "grad_norm": 0.578125, "learning_rate": 0.0004975825922369141, "loss": 5.2451, "mean_token_accuracy": 0.2112227827310562, "num_tokens": 24787728.0, "step": 1525 }, { "entropy": 5.326749706268311, "epoch": 1.251124744376278, "grad_norm": 0.57421875, "learning_rate": 0.0004975363231746052, "loss": 5.2057, "mean_token_accuracy": 0.21062490195035935, "num_tokens": 24869224.0, "step": 1530 }, { "entropy": 5.424345874786377, "epoch": 1.2552147239263804, "grad_norm": 0.5703125, "learning_rate": 0.000497489617943214, "loss": 5.2592, "mean_token_accuracy": 0.20984654426574706, "num_tokens": 24950396.0, "step": 1535 }, { "entropy": 5.402578926086425, "epoch": 1.2593047034764826, "grad_norm": 0.57421875, "learning_rate": 0.0004974424766342821, "loss": 5.2413, "mean_token_accuracy": 0.21081471145153047, "num_tokens": 25031493.0, "step": 1540 }, { "entropy": 5.384630012512207, "epoch": 1.263394683026585, "grad_norm": 0.55859375, "learning_rate": 0.0004973948993402065, "loss": 5.276, "mean_token_accuracy": 0.2086309775710106, "num_tokens": 25112998.0, "step": 1545 }, { "entropy": 5.413353204727173, "epoch": 1.2674846625766871, "grad_norm": 0.56640625, "learning_rate": 0.0004973468861542381, "loss": 5.286, "mean_token_accuracy": 0.2099459558725357, "num_tokens": 25194561.0, "step": 1550 }, { "entropy": 5.412050342559814, "epoch": 1.2715746421267893, "grad_norm": 0.5859375, "learning_rate": 0.0004972984371704819, "loss": 5.2451, "mean_token_accuracy": 0.21018154919147491, "num_tokens": 25276138.0, "step": 1555 }, { "entropy": 5.352128028869629, "epoch": 1.2756646216768917, "grad_norm": 0.6015625, "learning_rate": 0.0004972495524838979, "loss": 5.2298, "mean_token_accuracy": 0.21101472675800323, "num_tokens": 25357818.0, "step": 1560 }, { "entropy": 5.383926868438721, "epoch": 1.2797546012269938, "grad_norm": 0.61328125, "learning_rate": 0.0004972002321902994, "loss": 5.2599, "mean_token_accuracy": 0.2116800844669342, "num_tokens": 25439409.0, "step": 1565 }, { "entropy": 5.351431512832642, "epoch": 1.2838445807770962, "grad_norm": 0.58203125, "learning_rate": 0.0004971504763863539, "loss": 5.1761, "mean_token_accuracy": 0.21625618040561675, "num_tokens": 25521033.0, "step": 1570 }, { "entropy": 5.42213659286499, "epoch": 1.2879345603271983, "grad_norm": 0.59375, "learning_rate": 0.000497100285169582, "loss": 5.2741, "mean_token_accuracy": 0.20880339294672012, "num_tokens": 25602048.0, "step": 1575 }, { "entropy": 5.367815637588501, "epoch": 1.2920245398773007, "grad_norm": 0.5625, "learning_rate": 0.0004970496586383582, "loss": 5.2493, "mean_token_accuracy": 0.20914248377084732, "num_tokens": 25683633.0, "step": 1580 }, { "entropy": 5.3704719066619875, "epoch": 1.2961145194274029, "grad_norm": 0.5390625, "learning_rate": 0.0004969985968919102, "loss": 5.2486, "mean_token_accuracy": 0.20943263173103333, "num_tokens": 25764437.0, "step": 1585 }, { "entropy": 5.36567120552063, "epoch": 1.300204498977505, "grad_norm": 0.5546875, "learning_rate": 0.0004969471000303181, "loss": 5.2283, "mean_token_accuracy": 0.21231106370687486, "num_tokens": 25846060.0, "step": 1590 }, { "entropy": 5.389878988265991, "epoch": 1.3042944785276074, "grad_norm": 0.54296875, "learning_rate": 0.0004968951681545156, "loss": 5.2752, "mean_token_accuracy": 0.21034973859786987, "num_tokens": 25927577.0, "step": 1595 }, { "entropy": 5.3402313709259035, "epoch": 1.3083844580777095, "grad_norm": 0.56640625, "learning_rate": 0.0004968428013662885, "loss": 5.2171, "mean_token_accuracy": 0.20923437029123307, "num_tokens": 26009219.0, "step": 1600 }, { "entropy": 5.352266502380371, "epoch": 1.312474437627812, "grad_norm": 0.5859375, "learning_rate": 0.0004967899997682754, "loss": 5.2198, "mean_token_accuracy": 0.2122654438018799, "num_tokens": 26089950.0, "step": 1605 }, { "entropy": 5.377188014984131, "epoch": 1.316564417177914, "grad_norm": 0.56640625, "learning_rate": 0.0004967367634639669, "loss": 5.2322, "mean_token_accuracy": 0.2092613697052002, "num_tokens": 26171108.0, "step": 1610 }, { "entropy": 5.349839735031128, "epoch": 1.3206543967280164, "grad_norm": 0.60546875, "learning_rate": 0.0004966830925577054, "loss": 5.1826, "mean_token_accuracy": 0.21664625704288482, "num_tokens": 26251936.0, "step": 1615 }, { "entropy": 5.320872497558594, "epoch": 1.3247443762781186, "grad_norm": 0.60546875, "learning_rate": 0.0004966289871546856, "loss": 5.1904, "mean_token_accuracy": 0.21548739075660706, "num_tokens": 26333564.0, "step": 1620 }, { "entropy": 5.345857954025268, "epoch": 1.3288343558282207, "grad_norm": 0.55859375, "learning_rate": 0.0004965744473609536, "loss": 5.2017, "mean_token_accuracy": 0.2132759615778923, "num_tokens": 26415058.0, "step": 1625 }, { "entropy": 5.306171989440918, "epoch": 1.332924335378323, "grad_norm": 0.55078125, "learning_rate": 0.0004965194732834067, "loss": 5.1631, "mean_token_accuracy": 0.21727183908224107, "num_tokens": 26496004.0, "step": 1630 }, { "entropy": 5.361862277984619, "epoch": 1.3370143149284255, "grad_norm": 0.5703125, "learning_rate": 0.0004964640650297937, "loss": 5.2334, "mean_token_accuracy": 0.21221859753131866, "num_tokens": 26577670.0, "step": 1635 }, { "entropy": 5.352596378326416, "epoch": 1.3411042944785276, "grad_norm": 0.59375, "learning_rate": 0.0004964082227087141, "loss": 5.2155, "mean_token_accuracy": 0.21038477271795272, "num_tokens": 26659325.0, "step": 1640 }, { "entropy": 5.303871917724609, "epoch": 1.3451942740286298, "grad_norm": 0.546875, "learning_rate": 0.0004963519464296186, "loss": 5.1476, "mean_token_accuracy": 0.21471314132213593, "num_tokens": 26740994.0, "step": 1645 }, { "entropy": 5.371114063262939, "epoch": 1.3492842535787322, "grad_norm": 0.54296875, "learning_rate": 0.000496295236302808, "loss": 5.2177, "mean_token_accuracy": 0.2107516422867775, "num_tokens": 26821950.0, "step": 1650 }, { "entropy": 5.314587593078613, "epoch": 1.3533742331288343, "grad_norm": 0.5625, "learning_rate": 0.0004962380924394336, "loss": 5.1885, "mean_token_accuracy": 0.2120109036564827, "num_tokens": 26902484.0, "step": 1655 }, { "entropy": 5.33351411819458, "epoch": 1.3574642126789367, "grad_norm": 0.53515625, "learning_rate": 0.0004961805149514968, "loss": 5.2004, "mean_token_accuracy": 0.21120501756668092, "num_tokens": 26984109.0, "step": 1660 }, { "entropy": 5.370116758346557, "epoch": 1.3615541922290388, "grad_norm": 0.55078125, "learning_rate": 0.0004961225039518491, "loss": 5.2152, "mean_token_accuracy": 0.2139083132147789, "num_tokens": 27065441.0, "step": 1665 }, { "entropy": 5.316911840438843, "epoch": 1.3656441717791412, "grad_norm": 0.6015625, "learning_rate": 0.0004960640595541912, "loss": 5.1803, "mean_token_accuracy": 0.21333669424057006, "num_tokens": 27146917.0, "step": 1670 }, { "entropy": 5.315832185745239, "epoch": 1.3697341513292434, "grad_norm": 0.578125, "learning_rate": 0.0004960051818730739, "loss": 5.1623, "mean_token_accuracy": 0.21655652076005935, "num_tokens": 27228380.0, "step": 1675 }, { "entropy": 5.311358118057251, "epoch": 1.3738241308793455, "grad_norm": 0.5625, "learning_rate": 0.0004959458710238967, "loss": 5.1872, "mean_token_accuracy": 0.21530735939741136, "num_tokens": 27309878.0, "step": 1680 }, { "entropy": 5.279473972320557, "epoch": 1.377914110429448, "grad_norm": 0.6015625, "learning_rate": 0.0004958861271229083, "loss": 5.1339, "mean_token_accuracy": 0.21896741539239883, "num_tokens": 27391119.0, "step": 1685 }, { "entropy": 5.307453060150147, "epoch": 1.38200408997955, "grad_norm": 0.5546875, "learning_rate": 0.0004958259502872063, "loss": 5.1846, "mean_token_accuracy": 0.21518977135419845, "num_tokens": 27472585.0, "step": 1690 }, { "entropy": 5.310364723205566, "epoch": 1.3860940695296524, "grad_norm": 0.53125, "learning_rate": 0.0004957653406347368, "loss": 5.1847, "mean_token_accuracy": 0.21337068527936937, "num_tokens": 27554204.0, "step": 1695 }, { "entropy": 5.3163079738616945, "epoch": 1.3901840490797546, "grad_norm": 0.5390625, "learning_rate": 0.0004957042982842938, "loss": 5.2003, "mean_token_accuracy": 0.21462427079677582, "num_tokens": 27635818.0, "step": 1700 }, { "entropy": 5.325690841674804, "epoch": 1.394274028629857, "grad_norm": 0.57421875, "learning_rate": 0.0004956428233555201, "loss": 5.1614, "mean_token_accuracy": 0.2142445057630539, "num_tokens": 27717451.0, "step": 1705 }, { "entropy": 5.282151985168457, "epoch": 1.398364008179959, "grad_norm": 0.546875, "learning_rate": 0.0004955809159689056, "loss": 5.1659, "mean_token_accuracy": 0.21574875414371492, "num_tokens": 27799140.0, "step": 1710 }, { "entropy": 5.292479753494263, "epoch": 1.4024539877300612, "grad_norm": 0.53515625, "learning_rate": 0.0004955185762457885, "loss": 5.1761, "mean_token_accuracy": 0.21379369050264357, "num_tokens": 27880478.0, "step": 1715 }, { "entropy": 5.357530784606934, "epoch": 1.4065439672801636, "grad_norm": 0.59765625, "learning_rate": 0.000495455804308354, "loss": 5.2123, "mean_token_accuracy": 0.21430829018354416, "num_tokens": 27961534.0, "step": 1720 }, { "entropy": 5.3363627910614015, "epoch": 1.4106339468302658, "grad_norm": 0.5546875, "learning_rate": 0.0004953926002796345, "loss": 5.2226, "mean_token_accuracy": 0.21202680766582488, "num_tokens": 28042443.0, "step": 1725 }, { "entropy": 5.305199670791626, "epoch": 1.4147239263803681, "grad_norm": 0.546875, "learning_rate": 0.0004953289642835091, "loss": 5.1683, "mean_token_accuracy": 0.21437540650367737, "num_tokens": 28123466.0, "step": 1730 }, { "entropy": 5.300405883789063, "epoch": 1.4188139059304703, "grad_norm": 0.55859375, "learning_rate": 0.0004952648964447041, "loss": 5.1851, "mean_token_accuracy": 0.21557315140962602, "num_tokens": 28205143.0, "step": 1735 }, { "entropy": 5.277932786941529, "epoch": 1.4229038854805727, "grad_norm": 0.55859375, "learning_rate": 0.0004952003968887916, "loss": 5.1274, "mean_token_accuracy": 0.21827540248632432, "num_tokens": 28286769.0, "step": 1740 }, { "entropy": 5.284214162826538, "epoch": 1.4269938650306748, "grad_norm": 0.578125, "learning_rate": 0.0004951354657421901, "loss": 5.1766, "mean_token_accuracy": 0.21483704596757888, "num_tokens": 28368403.0, "step": 1745 }, { "entropy": 5.3313515186309814, "epoch": 1.431083844580777, "grad_norm": 0.55859375, "learning_rate": 0.0004950701031321642, "loss": 5.1947, "mean_token_accuracy": 0.2154526501893997, "num_tokens": 28449660.0, "step": 1750 }, { "entropy": 5.302390766143799, "epoch": 1.4351738241308793, "grad_norm": 0.55859375, "learning_rate": 0.0004950043091868239, "loss": 5.1842, "mean_token_accuracy": 0.21530397236347198, "num_tokens": 28530347.0, "step": 1755 }, { "entropy": 5.335675287246704, "epoch": 1.4392638036809817, "grad_norm": 0.5546875, "learning_rate": 0.0004949380840351248, "loss": 5.1848, "mean_token_accuracy": 0.2136605054140091, "num_tokens": 28611497.0, "step": 1760 }, { "entropy": 5.274587726593017, "epoch": 1.4433537832310839, "grad_norm": 0.53515625, "learning_rate": 0.0004948714278068674, "loss": 5.1551, "mean_token_accuracy": 0.21847329884767533, "num_tokens": 28692918.0, "step": 1765 }, { "entropy": 5.320195436477661, "epoch": 1.447443762781186, "grad_norm": 0.55078125, "learning_rate": 0.0004948043406326975, "loss": 5.2081, "mean_token_accuracy": 0.21080148369073867, "num_tokens": 28774231.0, "step": 1770 }, { "entropy": 5.31915512084961, "epoch": 1.4515337423312884, "grad_norm": 0.5234375, "learning_rate": 0.0004947368226441049, "loss": 5.1944, "mean_token_accuracy": 0.21410203874111175, "num_tokens": 28855173.0, "step": 1775 }, { "entropy": 5.322488594055176, "epoch": 1.4556237218813906, "grad_norm": 0.55859375, "learning_rate": 0.0004946688739734249, "loss": 5.1898, "mean_token_accuracy": 0.2166566178202629, "num_tokens": 28935963.0, "step": 1780 }, { "entropy": 5.294971990585327, "epoch": 1.4597137014314927, "grad_norm": 0.56640625, "learning_rate": 0.0004946004947538355, "loss": 5.154, "mean_token_accuracy": 0.2182640939950943, "num_tokens": 29017556.0, "step": 1785 }, { "entropy": 5.23001446723938, "epoch": 1.463803680981595, "grad_norm": 0.55859375, "learning_rate": 0.0004945316851193599, "loss": 5.1307, "mean_token_accuracy": 0.22145968526601792, "num_tokens": 29098366.0, "step": 1790 }, { "entropy": 5.295672416687012, "epoch": 1.4678936605316975, "grad_norm": 0.53515625, "learning_rate": 0.000494462445204864, "loss": 5.1644, "mean_token_accuracy": 0.21546489000320435, "num_tokens": 29179494.0, "step": 1795 }, { "entropy": 5.24140305519104, "epoch": 1.4719836400817996, "grad_norm": 0.54296875, "learning_rate": 0.0004943927751460575, "loss": 5.1103, "mean_token_accuracy": 0.22137825936079025, "num_tokens": 29260842.0, "step": 1800 }, { "entropy": 5.308050298690796, "epoch": 1.4760736196319018, "grad_norm": 0.53515625, "learning_rate": 0.0004943226750794931, "loss": 5.1512, "mean_token_accuracy": 0.219248366355896, "num_tokens": 29342266.0, "step": 1805 }, { "entropy": 5.260629463195801, "epoch": 1.4801635991820041, "grad_norm": 0.51171875, "learning_rate": 0.0004942521451425662, "loss": 5.1206, "mean_token_accuracy": 0.22339380234479905, "num_tokens": 29423714.0, "step": 1810 }, { "entropy": 5.263856554031372, "epoch": 1.4842535787321063, "grad_norm": 0.55859375, "learning_rate": 0.0004941811854735148, "loss": 5.1343, "mean_token_accuracy": 0.2196061447262764, "num_tokens": 29504916.0, "step": 1815 }, { "entropy": 5.210256147384643, "epoch": 1.4883435582822087, "grad_norm": 0.53515625, "learning_rate": 0.0004941097962114191, "loss": 5.0801, "mean_token_accuracy": 0.2189863935112953, "num_tokens": 29586370.0, "step": 1820 }, { "entropy": 5.2425439834594725, "epoch": 1.4924335378323108, "grad_norm": 0.5703125, "learning_rate": 0.0004940379774962016, "loss": 5.1143, "mean_token_accuracy": 0.21683698296546935, "num_tokens": 29667862.0, "step": 1825 }, { "entropy": 5.296167707443237, "epoch": 1.4965235173824132, "grad_norm": 0.54296875, "learning_rate": 0.0004939657294686262, "loss": 5.1769, "mean_token_accuracy": 0.2177164763212204, "num_tokens": 29749411.0, "step": 1830 }, { "entropy": 5.260873317718506, "epoch": 1.5006134969325153, "grad_norm": 0.5625, "learning_rate": 0.0004938930522702981, "loss": 5.1203, "mean_token_accuracy": 0.22224463969469072, "num_tokens": 29830012.0, "step": 1835 }, { "entropy": 5.287509298324585, "epoch": 1.5047034764826175, "grad_norm": 0.55859375, "learning_rate": 0.0004938199460436644, "loss": 5.1491, "mean_token_accuracy": 0.21774934977293015, "num_tokens": 29911099.0, "step": 1840 }, { "entropy": 5.271704816818238, "epoch": 1.5087934560327199, "grad_norm": 0.5390625, "learning_rate": 0.0004937464109320121, "loss": 5.1282, "mean_token_accuracy": 0.21638013273477555, "num_tokens": 29992708.0, "step": 1845 }, { "entropy": 5.204687547683716, "epoch": 1.5128834355828222, "grad_norm": 0.5390625, "learning_rate": 0.0004936724470794698, "loss": 5.0933, "mean_token_accuracy": 0.22308481931686402, "num_tokens": 30073684.0, "step": 1850 }, { "entropy": 5.280852222442627, "epoch": 1.5169734151329244, "grad_norm": 0.54296875, "learning_rate": 0.0004935980546310056, "loss": 5.1494, "mean_token_accuracy": 0.21621030420064927, "num_tokens": 30154985.0, "step": 1855 }, { "entropy": 5.265926933288574, "epoch": 1.5210633946830265, "grad_norm": 0.578125, "learning_rate": 0.0004935232337324281, "loss": 5.1298, "mean_token_accuracy": 0.218222077190876, "num_tokens": 30236055.0, "step": 1860 }, { "entropy": 5.181027126312256, "epoch": 1.525153374233129, "grad_norm": 0.53515625, "learning_rate": 0.0004934479845303855, "loss": 5.0798, "mean_token_accuracy": 0.22376650273799897, "num_tokens": 30317255.0, "step": 1865 }, { "entropy": 5.218193578720093, "epoch": 1.529243353783231, "grad_norm": 0.5390625, "learning_rate": 0.0004933723071723657, "loss": 5.0637, "mean_token_accuracy": 0.22125799655914308, "num_tokens": 30398875.0, "step": 1870 }, { "entropy": 5.2794976234436035, "epoch": 1.5333333333333332, "grad_norm": 0.52734375, "learning_rate": 0.0004932962018066955, "loss": 5.1213, "mean_token_accuracy": 0.21654312014579774, "num_tokens": 30479967.0, "step": 1875 }, { "entropy": 5.216181373596191, "epoch": 1.5374233128834356, "grad_norm": 0.54296875, "learning_rate": 0.0004932196685825406, "loss": 5.1177, "mean_token_accuracy": 0.21808333992958068, "num_tokens": 30561568.0, "step": 1880 }, { "entropy": 5.251797199249268, "epoch": 1.541513292433538, "grad_norm": 0.5390625, "learning_rate": 0.0004931427076499056, "loss": 5.1259, "mean_token_accuracy": 0.21767597943544387, "num_tokens": 30642499.0, "step": 1885 }, { "entropy": 5.201221704483032, "epoch": 1.5456032719836401, "grad_norm": 0.5625, "learning_rate": 0.000493065319159633, "loss": 5.0547, "mean_token_accuracy": 0.22425012290477753, "num_tokens": 30724177.0, "step": 1890 }, { "entropy": 5.228001832962036, "epoch": 1.5496932515337423, "grad_norm": 0.52734375, "learning_rate": 0.0004929875032634035, "loss": 5.1235, "mean_token_accuracy": 0.22026432305574417, "num_tokens": 30805262.0, "step": 1895 }, { "entropy": 5.223613691329956, "epoch": 1.5537832310838446, "grad_norm": 0.578125, "learning_rate": 0.0004929092601137357, "loss": 5.1095, "mean_token_accuracy": 0.22017054855823517, "num_tokens": 30885895.0, "step": 1900 }, { "entropy": 5.212532472610474, "epoch": 1.5578732106339468, "grad_norm": 0.5234375, "learning_rate": 0.0004928305898639852, "loss": 5.082, "mean_token_accuracy": 0.22250520586967468, "num_tokens": 30967129.0, "step": 1905 }, { "entropy": 5.251621246337891, "epoch": 1.561963190184049, "grad_norm": 0.5234375, "learning_rate": 0.0004927514926683453, "loss": 5.1259, "mean_token_accuracy": 0.21923950016498567, "num_tokens": 31047444.0, "step": 1910 }, { "entropy": 5.231003952026367, "epoch": 1.5660531697341513, "grad_norm": 0.5546875, "learning_rate": 0.0004926719686818454, "loss": 5.1226, "mean_token_accuracy": 0.2174128398299217, "num_tokens": 31128608.0, "step": 1915 }, { "entropy": 5.246667766571045, "epoch": 1.5701431492842537, "grad_norm": 0.55859375, "learning_rate": 0.0004925920180603519, "loss": 5.153, "mean_token_accuracy": 0.21650979071855544, "num_tokens": 31209876.0, "step": 1920 }, { "entropy": 5.202984809875488, "epoch": 1.5742331288343558, "grad_norm": 0.5546875, "learning_rate": 0.0004925116409605671, "loss": 5.0773, "mean_token_accuracy": 0.22348576337099074, "num_tokens": 31291022.0, "step": 1925 }, { "entropy": 5.20338454246521, "epoch": 1.578323108384458, "grad_norm": 0.5625, "learning_rate": 0.0004924308375400297, "loss": 5.1014, "mean_token_accuracy": 0.22186137288808822, "num_tokens": 31372453.0, "step": 1930 }, { "entropy": 5.26684045791626, "epoch": 1.5824130879345604, "grad_norm": 0.55078125, "learning_rate": 0.0004923496079571135, "loss": 5.1545, "mean_token_accuracy": 0.2170016273856163, "num_tokens": 31453152.0, "step": 1935 }, { "entropy": 5.261339616775513, "epoch": 1.5865030674846625, "grad_norm": 0.546875, "learning_rate": 0.0004922679523710274, "loss": 5.1304, "mean_token_accuracy": 0.21951664388179778, "num_tokens": 31534596.0, "step": 1940 }, { "entropy": 5.250435066223145, "epoch": 1.5905930470347647, "grad_norm": 0.5390625, "learning_rate": 0.000492185870941816, "loss": 5.0986, "mean_token_accuracy": 0.22095159590244293, "num_tokens": 31615614.0, "step": 1945 }, { "entropy": 5.228601551055908, "epoch": 1.594683026584867, "grad_norm": 0.5390625, "learning_rate": 0.0004921033638303578, "loss": 5.1165, "mean_token_accuracy": 0.21794263869524003, "num_tokens": 31696826.0, "step": 1950 }, { "entropy": 5.260002613067627, "epoch": 1.5987730061349694, "grad_norm": 0.56640625, "learning_rate": 0.0004920204311983662, "loss": 5.1288, "mean_token_accuracy": 0.21532346457242965, "num_tokens": 31778089.0, "step": 1955 }, { "entropy": 5.198488521575928, "epoch": 1.6028629856850716, "grad_norm": 0.5234375, "learning_rate": 0.0004919370732083884, "loss": 5.0856, "mean_token_accuracy": 0.22283379286527633, "num_tokens": 31859134.0, "step": 1960 }, { "entropy": 5.277430057525635, "epoch": 1.6069529652351737, "grad_norm": 0.515625, "learning_rate": 0.0004918532900238049, "loss": 5.1458, "mean_token_accuracy": 0.21547970175743103, "num_tokens": 31940389.0, "step": 1965 }, { "entropy": 5.197055435180664, "epoch": 1.611042944785276, "grad_norm": 0.546875, "learning_rate": 0.0004917690818088302, "loss": 5.0996, "mean_token_accuracy": 0.22084179073572158, "num_tokens": 32021558.0, "step": 1970 }, { "entropy": 5.254745626449585, "epoch": 1.6151329243353785, "grad_norm": 0.55078125, "learning_rate": 0.0004916844487285117, "loss": 5.1115, "mean_token_accuracy": 0.22150487452745438, "num_tokens": 32103199.0, "step": 1975 }, { "entropy": 5.208833503723144, "epoch": 1.6192229038854806, "grad_norm": 0.52734375, "learning_rate": 0.0004915993909487293, "loss": 5.0712, "mean_token_accuracy": 0.22096495181322098, "num_tokens": 32184719.0, "step": 1980 }, { "entropy": 5.257219648361206, "epoch": 1.6233128834355828, "grad_norm": 0.56640625, "learning_rate": 0.0004915139086361955, "loss": 5.1384, "mean_token_accuracy": 0.21668985933065416, "num_tokens": 32266013.0, "step": 1985 }, { "entropy": 5.269886636734009, "epoch": 1.6274028629856852, "grad_norm": 0.53125, "learning_rate": 0.0004914280019584548, "loss": 5.1286, "mean_token_accuracy": 0.21631538718938828, "num_tokens": 32346865.0, "step": 1990 }, { "entropy": 5.21401424407959, "epoch": 1.6314928425357873, "grad_norm": 0.5625, "learning_rate": 0.0004913416710838834, "loss": 5.0861, "mean_token_accuracy": 0.22043754309415817, "num_tokens": 32427435.0, "step": 1995 }, { "entropy": 5.179972743988037, "epoch": 1.6355828220858895, "grad_norm": 0.53515625, "learning_rate": 0.0004912549161816891, "loss": 5.0324, "mean_token_accuracy": 0.22298528850078583, "num_tokens": 32508028.0, "step": 2000 }, { "epoch": 1.6355828220858895, "eval_entropy": 5.060819897412333, "eval_loss": 5.099005699157715, "eval_mean_token_accuracy": 0.22219983919822112, "eval_num_tokens": 32508028.0, "eval_runtime": 68.2032, "eval_samples_per_second": 60.657, "eval_steps_per_second": 7.595, "step": 2000 }, { "entropy": 5.230832099914551, "epoch": 1.6396728016359918, "grad_norm": 0.55078125, "learning_rate": 0.0004911677374219106, "loss": 5.1007, "mean_token_accuracy": 0.219247205555439, "num_tokens": 32589616.0, "step": 2005 }, { "entropy": 5.282004928588867, "epoch": 1.6437627811860942, "grad_norm": 0.5546875, "learning_rate": 0.0004910801349754176, "loss": 5.1599, "mean_token_accuracy": 0.2180356949567795, "num_tokens": 32671151.0, "step": 2010 }, { "entropy": 5.2055747509002686, "epoch": 1.6478527607361964, "grad_norm": 0.5234375, "learning_rate": 0.0004909921090139099, "loss": 5.1002, "mean_token_accuracy": 0.21658260375261307, "num_tokens": 32752807.0, "step": 2015 }, { "entropy": 5.28322925567627, "epoch": 1.6519427402862985, "grad_norm": 0.51953125, "learning_rate": 0.0004909036597099174, "loss": 5.1589, "mean_token_accuracy": 0.21561679095029831, "num_tokens": 32833617.0, "step": 2020 }, { "entropy": 5.2126072406768795, "epoch": 1.6560327198364009, "grad_norm": 0.55078125, "learning_rate": 0.0004908147872368004, "loss": 5.1056, "mean_token_accuracy": 0.21733339577913285, "num_tokens": 32914877.0, "step": 2025 }, { "entropy": 5.1951744556427, "epoch": 1.660122699386503, "grad_norm": 0.546875, "learning_rate": 0.0004907254917687476, "loss": 5.0534, "mean_token_accuracy": 0.22132280319929123, "num_tokens": 32996447.0, "step": 2030 }, { "entropy": 5.219225740432739, "epoch": 1.6642126789366052, "grad_norm": 0.546875, "learning_rate": 0.0004906357734807776, "loss": 5.1046, "mean_token_accuracy": 0.2222513198852539, "num_tokens": 33077697.0, "step": 2035 }, { "entropy": 5.245347261428833, "epoch": 1.6683026584867076, "grad_norm": 0.54296875, "learning_rate": 0.0004905456325487371, "loss": 5.1165, "mean_token_accuracy": 0.217647522687912, "num_tokens": 33159276.0, "step": 2040 }, { "entropy": 5.200337696075439, "epoch": 1.67239263803681, "grad_norm": 0.53515625, "learning_rate": 0.0004904550691493017, "loss": 5.0576, "mean_token_accuracy": 0.22400822639465331, "num_tokens": 33240967.0, "step": 2045 }, { "entropy": 5.2077075958251955, "epoch": 1.676482617586912, "grad_norm": 0.5234375, "learning_rate": 0.0004903640834599748, "loss": 5.1019, "mean_token_accuracy": 0.21800609529018403, "num_tokens": 33322294.0, "step": 2050 }, { "entropy": 5.163883352279663, "epoch": 1.6805725971370142, "grad_norm": 0.55859375, "learning_rate": 0.0004902726756590874, "loss": 5.0364, "mean_token_accuracy": 0.22468381524085998, "num_tokens": 33403314.0, "step": 2055 }, { "entropy": 5.231391286849975, "epoch": 1.6846625766871166, "grad_norm": 0.53125, "learning_rate": 0.0004901808459257979, "loss": 5.0929, "mean_token_accuracy": 0.21906571984291076, "num_tokens": 33484476.0, "step": 2060 }, { "entropy": 5.200314474105835, "epoch": 1.6887525562372188, "grad_norm": 0.578125, "learning_rate": 0.0004900885944400918, "loss": 5.1042, "mean_token_accuracy": 0.2202153757214546, "num_tokens": 33565557.0, "step": 2065 }, { "entropy": 5.202909088134765, "epoch": 1.692842535787321, "grad_norm": 0.5234375, "learning_rate": 0.0004899959213827811, "loss": 5.0653, "mean_token_accuracy": 0.22281403094530106, "num_tokens": 33646028.0, "step": 2070 }, { "entropy": 5.201122522354126, "epoch": 1.6969325153374233, "grad_norm": 0.5546875, "learning_rate": 0.0004899028269355041, "loss": 5.0843, "mean_token_accuracy": 0.21973094940185547, "num_tokens": 33727473.0, "step": 2075 }, { "entropy": 5.225057315826416, "epoch": 1.7010224948875257, "grad_norm": 0.53125, "learning_rate": 0.0004898093112807251, "loss": 5.083, "mean_token_accuracy": 0.22011276185512543, "num_tokens": 33807948.0, "step": 2080 }, { "entropy": 5.175272464752197, "epoch": 1.7051124744376278, "grad_norm": 0.546875, "learning_rate": 0.0004897153746017337, "loss": 5.0544, "mean_token_accuracy": 0.22442440688610077, "num_tokens": 33888885.0, "step": 2085 }, { "entropy": 5.156661558151245, "epoch": 1.70920245398773, "grad_norm": 0.54296875, "learning_rate": 0.0004896210170826451, "loss": 5.0425, "mean_token_accuracy": 0.22324688732624054, "num_tokens": 33970560.0, "step": 2090 }, { "entropy": 5.175463390350342, "epoch": 1.7132924335378323, "grad_norm": 0.59765625, "learning_rate": 0.0004895262389083991, "loss": 5.0418, "mean_token_accuracy": 0.22385067194700242, "num_tokens": 34051820.0, "step": 2095 }, { "entropy": 5.209377145767212, "epoch": 1.7173824130879347, "grad_norm": 0.578125, "learning_rate": 0.00048943104026476, "loss": 5.0929, "mean_token_accuracy": 0.2208719104528427, "num_tokens": 34133460.0, "step": 2100 }, { "entropy": 5.161018514633179, "epoch": 1.7214723926380369, "grad_norm": 0.53515625, "learning_rate": 0.0004893354213383162, "loss": 5.0311, "mean_token_accuracy": 0.22417810410261155, "num_tokens": 34215000.0, "step": 2105 }, { "entropy": 5.242545080184937, "epoch": 1.725562372188139, "grad_norm": 0.51171875, "learning_rate": 0.0004892393823164799, "loss": 5.1095, "mean_token_accuracy": 0.21974032521247863, "num_tokens": 34296570.0, "step": 2110 }, { "entropy": 5.164000844955444, "epoch": 1.7296523517382414, "grad_norm": 0.52734375, "learning_rate": 0.0004891429233874867, "loss": 5.0566, "mean_token_accuracy": 0.22192672789096832, "num_tokens": 34377973.0, "step": 2115 }, { "entropy": 5.25903549194336, "epoch": 1.7337423312883435, "grad_norm": 0.53125, "learning_rate": 0.0004890460447403952, "loss": 5.1194, "mean_token_accuracy": 0.22196876108646393, "num_tokens": 34459589.0, "step": 2120 }, { "entropy": 5.181925201416016, "epoch": 1.7378323108384457, "grad_norm": 0.546875, "learning_rate": 0.0004889487465650867, "loss": 5.0919, "mean_token_accuracy": 0.22111362516880034, "num_tokens": 34540794.0, "step": 2125 }, { "entropy": 5.20899510383606, "epoch": 1.741922290388548, "grad_norm": 0.546875, "learning_rate": 0.0004888510290522645, "loss": 5.0524, "mean_token_accuracy": 0.22233040779829025, "num_tokens": 34622378.0, "step": 2130 }, { "entropy": 5.144508934020996, "epoch": 1.7460122699386504, "grad_norm": 0.54296875, "learning_rate": 0.0004887528923934542, "loss": 5.0402, "mean_token_accuracy": 0.22469659000635148, "num_tokens": 34703058.0, "step": 2135 }, { "entropy": 5.17820987701416, "epoch": 1.7501022494887526, "grad_norm": 0.55859375, "learning_rate": 0.0004886543367810026, "loss": 5.073, "mean_token_accuracy": 0.2250461384654045, "num_tokens": 34784514.0, "step": 2140 }, { "entropy": 5.207077932357788, "epoch": 1.7541922290388547, "grad_norm": 0.5390625, "learning_rate": 0.0004885553624080778, "loss": 5.1166, "mean_token_accuracy": 0.22008563280105592, "num_tokens": 34865237.0, "step": 2145 }, { "entropy": 5.218975162506103, "epoch": 1.7582822085889571, "grad_norm": 0.5390625, "learning_rate": 0.0004884559694686686, "loss": 5.0977, "mean_token_accuracy": 0.22253641486167908, "num_tokens": 34946689.0, "step": 2150 }, { "entropy": 5.185157775878906, "epoch": 1.7623721881390593, "grad_norm": 0.51953125, "learning_rate": 0.0004883561581575843, "loss": 5.0731, "mean_token_accuracy": 0.22343298345804213, "num_tokens": 35027182.0, "step": 2155 }, { "entropy": 5.185508966445923, "epoch": 1.7664621676891614, "grad_norm": 0.515625, "learning_rate": 0.00048825592867045406, "loss": 5.0593, "mean_token_accuracy": 0.22386937439441681, "num_tokens": 35108547.0, "step": 2160 }, { "entropy": 5.1671911716461185, "epoch": 1.7705521472392638, "grad_norm": 0.51171875, "learning_rate": 0.00048815528120372664, "loss": 5.04, "mean_token_accuracy": 0.22675409764051438, "num_tokens": 35189766.0, "step": 2165 }, { "entropy": 5.144899845123291, "epoch": 1.7746421267893662, "grad_norm": 0.53515625, "learning_rate": 0.0004880542159546703, "loss": 5.0022, "mean_token_accuracy": 0.2271200194954872, "num_tokens": 35271359.0, "step": 2170 }, { "entropy": 5.160240221023559, "epoch": 1.7787321063394683, "grad_norm": 0.5390625, "learning_rate": 0.00048795273312137187, "loss": 5.0417, "mean_token_accuracy": 0.22622190415859222, "num_tokens": 35352504.0, "step": 2175 }, { "entropy": 5.209652757644653, "epoch": 1.7828220858895705, "grad_norm": 0.53515625, "learning_rate": 0.00048785083290273676, "loss": 5.0668, "mean_token_accuracy": 0.2230725333094597, "num_tokens": 35434131.0, "step": 2180 }, { "entropy": 5.158541345596314, "epoch": 1.7869120654396728, "grad_norm": 0.53125, "learning_rate": 0.00048774851549848844, "loss": 5.0512, "mean_token_accuracy": 0.22355219721794128, "num_tokens": 35515745.0, "step": 2185 }, { "entropy": 5.168401050567627, "epoch": 1.791002044989775, "grad_norm": 0.53125, "learning_rate": 0.000487645781109168, "loss": 5.0502, "mean_token_accuracy": 0.22172819823026657, "num_tokens": 35597383.0, "step": 2190 }, { "entropy": 5.140007543563843, "epoch": 1.7950920245398772, "grad_norm": 0.54296875, "learning_rate": 0.0004875426299361339, "loss": 4.9949, "mean_token_accuracy": 0.22681471854448318, "num_tokens": 35678546.0, "step": 2195 }, { "entropy": 5.167155981063843, "epoch": 1.7991820040899795, "grad_norm": 0.55078125, "learning_rate": 0.0004874390621815615, "loss": 5.0232, "mean_token_accuracy": 0.2239782452583313, "num_tokens": 35760176.0, "step": 2200 }, { "entropy": 5.163224744796753, "epoch": 1.803271983640082, "grad_norm": 0.5234375, "learning_rate": 0.0004873350780484426, "loss": 5.0502, "mean_token_accuracy": 0.22481489926576614, "num_tokens": 35841579.0, "step": 2205 }, { "entropy": 5.166739988327026, "epoch": 1.807361963190184, "grad_norm": 0.51953125, "learning_rate": 0.00048723067774058507, "loss": 5.0334, "mean_token_accuracy": 0.22291862070560456, "num_tokens": 35923238.0, "step": 2210 }, { "entropy": 5.166903018951416, "epoch": 1.8114519427402862, "grad_norm": 0.50390625, "learning_rate": 0.00048712586146261254, "loss": 4.9943, "mean_token_accuracy": 0.22427679002285003, "num_tokens": 36004879.0, "step": 2215 }, { "entropy": 5.11282901763916, "epoch": 1.8155419222903886, "grad_norm": 0.5546875, "learning_rate": 0.00048702062941996397, "loss": 5.0456, "mean_token_accuracy": 0.22450873702764512, "num_tokens": 36085802.0, "step": 2220 }, { "entropy": 5.181997013092041, "epoch": 1.819631901840491, "grad_norm": 0.5625, "learning_rate": 0.0004869149818188931, "loss": 5.0397, "mean_token_accuracy": 0.2244134485721588, "num_tokens": 36167488.0, "step": 2225 }, { "entropy": 5.1066615104675295, "epoch": 1.823721881390593, "grad_norm": 0.55859375, "learning_rate": 0.00048680891886646824, "loss": 4.9839, "mean_token_accuracy": 0.22766097486019135, "num_tokens": 36249140.0, "step": 2230 }, { "entropy": 5.145777416229248, "epoch": 1.8278118609406953, "grad_norm": 0.5234375, "learning_rate": 0.0004867024407705717, "loss": 5.0302, "mean_token_accuracy": 0.22254575788974762, "num_tokens": 36330556.0, "step": 2235 }, { "entropy": 5.121987438201904, "epoch": 1.8319018404907976, "grad_norm": 0.52734375, "learning_rate": 0.00048659554773989975, "loss": 4.9884, "mean_token_accuracy": 0.22623222470283508, "num_tokens": 36411841.0, "step": 2240 }, { "entropy": 5.094103860855102, "epoch": 1.8359918200408998, "grad_norm": 0.53125, "learning_rate": 0.0004864882399839616, "loss": 4.9606, "mean_token_accuracy": 0.2304743632674217, "num_tokens": 36493457.0, "step": 2245 }, { "entropy": 5.173473787307739, "epoch": 1.840081799591002, "grad_norm": 0.51171875, "learning_rate": 0.00048638051771307934, "loss": 5.0328, "mean_token_accuracy": 0.22411035001277924, "num_tokens": 36574704.0, "step": 2250 }, { "entropy": 5.116518592834472, "epoch": 1.8441717791411043, "grad_norm": 0.52734375, "learning_rate": 0.0004862723811383877, "loss": 4.9922, "mean_token_accuracy": 0.2274567574262619, "num_tokens": 36656344.0, "step": 2255 }, { "entropy": 5.17039041519165, "epoch": 1.8482617586912067, "grad_norm": 0.52734375, "learning_rate": 0.0004861638304718334, "loss": 5.0681, "mean_token_accuracy": 0.22185750752687455, "num_tokens": 36737937.0, "step": 2260 }, { "entropy": 5.102982854843139, "epoch": 1.8523517382413088, "grad_norm": 0.52734375, "learning_rate": 0.00048605486592617455, "loss": 5.0033, "mean_token_accuracy": 0.22432126253843307, "num_tokens": 36819469.0, "step": 2265 }, { "entropy": 5.17413215637207, "epoch": 1.856441717791411, "grad_norm": 0.5234375, "learning_rate": 0.0004859454877149807, "loss": 5.0506, "mean_token_accuracy": 0.22338050305843354, "num_tokens": 36901120.0, "step": 2270 }, { "entropy": 5.086263990402221, "epoch": 1.8605316973415134, "grad_norm": 0.53515625, "learning_rate": 0.00048583569605263214, "loss": 4.9623, "mean_token_accuracy": 0.22979038953781128, "num_tokens": 36982741.0, "step": 2275 }, { "entropy": 5.119894313812256, "epoch": 1.8646216768916155, "grad_norm": 0.515625, "learning_rate": 0.00048572549115431953, "loss": 4.9717, "mean_token_accuracy": 0.22927713841199876, "num_tokens": 37064369.0, "step": 2280 }, { "entropy": 5.132665300369263, "epoch": 1.8687116564417177, "grad_norm": 0.51171875, "learning_rate": 0.00048561487323604325, "loss": 5.029, "mean_token_accuracy": 0.2232514277100563, "num_tokens": 37145574.0, "step": 2285 }, { "entropy": 5.108520460128784, "epoch": 1.87280163599182, "grad_norm": 0.52734375, "learning_rate": 0.00048550384251461355, "loss": 4.9801, "mean_token_accuracy": 0.2286642536520958, "num_tokens": 37226751.0, "step": 2290 }, { "entropy": 5.123102426528931, "epoch": 1.8768916155419224, "grad_norm": 0.5234375, "learning_rate": 0.0004853923992076495, "loss": 4.9668, "mean_token_accuracy": 0.2282184898853302, "num_tokens": 37308374.0, "step": 2295 }, { "entropy": 5.117383623123169, "epoch": 1.8809815950920246, "grad_norm": 0.51953125, "learning_rate": 0.0004852805435335789, "loss": 5.0001, "mean_token_accuracy": 0.22392120957374573, "num_tokens": 37390040.0, "step": 2300 }, { "entropy": 5.137595319747925, "epoch": 1.8850715746421267, "grad_norm": 0.54296875, "learning_rate": 0.00048516827571163794, "loss": 5.0233, "mean_token_accuracy": 0.22525953948497773, "num_tokens": 37471342.0, "step": 2305 }, { "entropy": 5.163462400436401, "epoch": 1.889161554192229, "grad_norm": 0.53125, "learning_rate": 0.00048505559596187037, "loss": 5.0456, "mean_token_accuracy": 0.2260495126247406, "num_tokens": 37552766.0, "step": 2310 }, { "entropy": 5.077380275726318, "epoch": 1.8932515337423312, "grad_norm": 0.5390625, "learning_rate": 0.0004849425045051275, "loss": 4.9646, "mean_token_accuracy": 0.22884069979190827, "num_tokens": 37633521.0, "step": 2315 }, { "entropy": 5.120154142379761, "epoch": 1.8973415132924334, "grad_norm": 0.546875, "learning_rate": 0.00048482900156306743, "loss": 5.0244, "mean_token_accuracy": 0.22485224455595015, "num_tokens": 37714517.0, "step": 2320 }, { "entropy": 5.148172855377197, "epoch": 1.9014314928425358, "grad_norm": 0.5390625, "learning_rate": 0.00048471508735815487, "loss": 4.981, "mean_token_accuracy": 0.23100253492593764, "num_tokens": 37796179.0, "step": 2325 }, { "entropy": 5.09586534500122, "epoch": 1.9055214723926381, "grad_norm": 0.515625, "learning_rate": 0.00048460076211366066, "loss": 5.0181, "mean_token_accuracy": 0.22343694120645524, "num_tokens": 37877864.0, "step": 2330 }, { "entropy": 5.144748067855835, "epoch": 1.9096114519427403, "grad_norm": 0.53515625, "learning_rate": 0.0004844860260536611, "loss": 5.0239, "mean_token_accuracy": 0.22540537118911744, "num_tokens": 37959398.0, "step": 2335 }, { "entropy": 5.170261573791504, "epoch": 1.9137014314928424, "grad_norm": 0.5546875, "learning_rate": 0.0004843708794030379, "loss": 5.0439, "mean_token_accuracy": 0.22846219390630723, "num_tokens": 38041015.0, "step": 2340 }, { "entropy": 5.173274898529053, "epoch": 1.9177914110429448, "grad_norm": 0.52734375, "learning_rate": 0.00048425532238747723, "loss": 5.0227, "mean_token_accuracy": 0.2285202592611313, "num_tokens": 38122114.0, "step": 2345 }, { "entropy": 5.082564163208008, "epoch": 1.9218813905930472, "grad_norm": 0.52734375, "learning_rate": 0.00048413935523346983, "loss": 5.0174, "mean_token_accuracy": 0.22781577259302138, "num_tokens": 38202958.0, "step": 2350 }, { "entropy": 5.094378709793091, "epoch": 1.9259713701431491, "grad_norm": 0.5078125, "learning_rate": 0.0004840229781683103, "loss": 4.9511, "mean_token_accuracy": 0.23180187940597535, "num_tokens": 38284314.0, "step": 2355 }, { "entropy": 5.074478960037231, "epoch": 1.9300613496932515, "grad_norm": 0.5234375, "learning_rate": 0.00048390619142009645, "loss": 4.9224, "mean_token_accuracy": 0.2326490893959999, "num_tokens": 38366006.0, "step": 2360 }, { "entropy": 5.124407482147217, "epoch": 1.9341513292433539, "grad_norm": 0.51171875, "learning_rate": 0.0004837889952177293, "loss": 5.0095, "mean_token_accuracy": 0.22925476282835006, "num_tokens": 38447585.0, "step": 2365 }, { "entropy": 5.148251914978028, "epoch": 1.938241308793456, "grad_norm": 0.53125, "learning_rate": 0.00048367138979091225, "loss": 5.0044, "mean_token_accuracy": 0.22654918134212493, "num_tokens": 38529281.0, "step": 2370 }, { "entropy": 5.131524181365966, "epoch": 1.9423312883435582, "grad_norm": 0.515625, "learning_rate": 0.0004835533753701508, "loss": 4.9803, "mean_token_accuracy": 0.22746318578720093, "num_tokens": 38610944.0, "step": 2375 }, { "entropy": 5.092132425308227, "epoch": 1.9464212678936605, "grad_norm": 0.55859375, "learning_rate": 0.00048343495218675217, "loss": 5.0067, "mean_token_accuracy": 0.2282467782497406, "num_tokens": 38692434.0, "step": 2380 }, { "entropy": 5.114362382888794, "epoch": 1.950511247443763, "grad_norm": 0.5078125, "learning_rate": 0.00048331612047282455, "loss": 4.9454, "mean_token_accuracy": 0.22840079963207244, "num_tokens": 38773726.0, "step": 2385 }, { "entropy": 5.071242809295654, "epoch": 1.954601226993865, "grad_norm": 0.5078125, "learning_rate": 0.00048319688046127716, "loss": 4.955, "mean_token_accuracy": 0.23158793598413469, "num_tokens": 38855082.0, "step": 2390 }, { "entropy": 5.076805830001831, "epoch": 1.9586912065439672, "grad_norm": 0.515625, "learning_rate": 0.000483077232385819, "loss": 4.9512, "mean_token_accuracy": 0.23025995939970018, "num_tokens": 38936304.0, "step": 2395 }, { "entropy": 5.095173120498657, "epoch": 1.9627811860940696, "grad_norm": 0.5078125, "learning_rate": 0.0004829571764809595, "loss": 4.9876, "mean_token_accuracy": 0.22899163365364075, "num_tokens": 39017954.0, "step": 2400 }, { "entropy": 5.125281238555909, "epoch": 1.9668711656441717, "grad_norm": 0.51953125, "learning_rate": 0.0004828367129820069, "loss": 4.9771, "mean_token_accuracy": 0.22636304199695587, "num_tokens": 39099352.0, "step": 2405 }, { "entropy": 5.077520656585693, "epoch": 1.970961145194274, "grad_norm": 0.51953125, "learning_rate": 0.00048271584212506856, "loss": 4.9657, "mean_token_accuracy": 0.22973643243312836, "num_tokens": 39180926.0, "step": 2410 }, { "entropy": 5.144113206863404, "epoch": 1.9750511247443763, "grad_norm": 0.54296875, "learning_rate": 0.0004825945641470502, "loss": 4.9833, "mean_token_accuracy": 0.22830794900655746, "num_tokens": 39262431.0, "step": 2415 }, { "entropy": 5.100745725631714, "epoch": 1.9791411042944786, "grad_norm": 0.53125, "learning_rate": 0.0004824728792856555, "loss": 4.9888, "mean_token_accuracy": 0.22754901498556138, "num_tokens": 39344021.0, "step": 2420 }, { "entropy": 5.130942106246948, "epoch": 1.9832310838445808, "grad_norm": 0.52734375, "learning_rate": 0.00048235078777938566, "loss": 4.9994, "mean_token_accuracy": 0.226763579249382, "num_tokens": 39425432.0, "step": 2425 }, { "entropy": 5.1207653999328615, "epoch": 1.987321063394683, "grad_norm": 0.51171875, "learning_rate": 0.00048222828986753884, "loss": 4.9986, "mean_token_accuracy": 0.2266847535967827, "num_tokens": 39505835.0, "step": 2430 }, { "entropy": 5.11164698600769, "epoch": 1.9914110429447853, "grad_norm": 0.515625, "learning_rate": 0.00048210538579020987, "loss": 5.0234, "mean_token_accuracy": 0.22876932173967363, "num_tokens": 39587482.0, "step": 2435 }, { "entropy": 5.114878988265991, "epoch": 1.9955010224948875, "grad_norm": 0.5234375, "learning_rate": 0.0004819820757882894, "loss": 4.9881, "mean_token_accuracy": 0.2273182362318039, "num_tokens": 39668397.0, "step": 2440 }, { "entropy": 5.141272830963135, "epoch": 1.9995910020449896, "grad_norm": 0.55859375, "learning_rate": 0.0004818583601034641, "loss": 5.021, "mean_token_accuracy": 0.22597697824239732, "num_tokens": 39749997.0, "step": 2445 }, { "entropy": 5.088277657826741, "epoch": 2.0032719836400816, "grad_norm": 0.51953125, "learning_rate": 0.0004817342389782153, "loss": 4.9387, "mean_token_accuracy": 0.2317948275142246, "num_tokens": 39815703.0, "step": 2450 }, { "entropy": 5.101982831954956, "epoch": 2.007361963190184, "grad_norm": 0.515625, "learning_rate": 0.00048160971265581937, "loss": 4.8676, "mean_token_accuracy": 0.23590912222862243, "num_tokens": 39897126.0, "step": 2455 }, { "entropy": 5.046277284622192, "epoch": 2.0114519427402864, "grad_norm": 0.5390625, "learning_rate": 0.0004814847813803467, "loss": 4.8438, "mean_token_accuracy": 0.23306371867656708, "num_tokens": 39977795.0, "step": 2460 }, { "entropy": 5.106733989715576, "epoch": 2.0155419222903888, "grad_norm": 0.546875, "learning_rate": 0.00048135944539666144, "loss": 4.8709, "mean_token_accuracy": 0.23460340201854707, "num_tokens": 40058962.0, "step": 2465 }, { "entropy": 5.082729530334473, "epoch": 2.0196319018404907, "grad_norm": 0.53515625, "learning_rate": 0.00048123370495042086, "loss": 4.8539, "mean_token_accuracy": 0.23306709676980972, "num_tokens": 40140240.0, "step": 2470 }, { "entropy": 5.101282548904419, "epoch": 2.023721881390593, "grad_norm": 0.53125, "learning_rate": 0.00048110756028807504, "loss": 4.8676, "mean_token_accuracy": 0.23361849784851074, "num_tokens": 40221855.0, "step": 2475 }, { "entropy": 5.059933280944824, "epoch": 2.0278118609406954, "grad_norm": 0.515625, "learning_rate": 0.00048098101165686634, "loss": 4.829, "mean_token_accuracy": 0.2366450011730194, "num_tokens": 40303294.0, "step": 2480 }, { "entropy": 5.062244319915772, "epoch": 2.0319018404907974, "grad_norm": 0.5234375, "learning_rate": 0.00048085405930482897, "loss": 4.8393, "mean_token_accuracy": 0.2368098795413971, "num_tokens": 40384732.0, "step": 2485 }, { "entropy": 5.0699244976043705, "epoch": 2.0359918200408997, "grad_norm": 0.53125, "learning_rate": 0.00048072670348078817, "loss": 4.8108, "mean_token_accuracy": 0.2379382997751236, "num_tokens": 40466361.0, "step": 2490 }, { "entropy": 5.018307065963745, "epoch": 2.040081799591002, "grad_norm": 0.5078125, "learning_rate": 0.00048059894443436025, "loss": 4.8096, "mean_token_accuracy": 0.23883476108312607, "num_tokens": 40547095.0, "step": 2495 }, { "entropy": 5.11606798171997, "epoch": 2.0441717791411045, "grad_norm": 0.51953125, "learning_rate": 0.00048047078241595173, "loss": 4.8865, "mean_token_accuracy": 0.2323901653289795, "num_tokens": 40628075.0, "step": 2500 }, { "epoch": 2.0441717791411045, "eval_entropy": 4.927756088580864, "eval_loss": 4.977355480194092, "eval_mean_token_accuracy": 0.23151352259413155, "eval_num_tokens": 40628075.0, "eval_runtime": 67.2266, "eval_samples_per_second": 61.538, "eval_steps_per_second": 7.705, "step": 2500 }, { "entropy": 5.012618637084961, "epoch": 2.0482617586912064, "grad_norm": 0.515625, "learning_rate": 0.00048034221767675897, "loss": 4.8093, "mean_token_accuracy": 0.2381066158413887, "num_tokens": 40709650.0, "step": 2505 }, { "entropy": 5.0944665431976315, "epoch": 2.052351738241309, "grad_norm": 0.49609375, "learning_rate": 0.00048021325046876753, "loss": 4.8655, "mean_token_accuracy": 0.23529266566038132, "num_tokens": 40791005.0, "step": 2510 }, { "entropy": 5.027026033401489, "epoch": 2.056441717791411, "grad_norm": 0.546875, "learning_rate": 0.00048008388104475193, "loss": 4.8039, "mean_token_accuracy": 0.23854458183050156, "num_tokens": 40872224.0, "step": 2515 }, { "entropy": 5.059017658233643, "epoch": 2.060531697341513, "grad_norm": 0.515625, "learning_rate": 0.00047995410965827513, "loss": 4.8419, "mean_token_accuracy": 0.2357902243733406, "num_tokens": 40953811.0, "step": 2520 }, { "entropy": 5.087168884277344, "epoch": 2.0646216768916155, "grad_norm": 0.53125, "learning_rate": 0.0004798239365636877, "loss": 4.8646, "mean_token_accuracy": 0.23564884960651397, "num_tokens": 41035410.0, "step": 2525 }, { "entropy": 5.075830030441284, "epoch": 2.068711656441718, "grad_norm": 0.515625, "learning_rate": 0.0004796933620161278, "loss": 4.8588, "mean_token_accuracy": 0.23574852496385573, "num_tokens": 41116885.0, "step": 2530 }, { "entropy": 5.035322380065918, "epoch": 2.07280163599182, "grad_norm": 0.5, "learning_rate": 0.00047956238627152014, "loss": 4.8018, "mean_token_accuracy": 0.2393171101808548, "num_tokens": 41198593.0, "step": 2535 }, { "entropy": 5.051853799819947, "epoch": 2.076891615541922, "grad_norm": 0.515625, "learning_rate": 0.0004794310095865761, "loss": 4.8514, "mean_token_accuracy": 0.23528196662664413, "num_tokens": 41280227.0, "step": 2540 }, { "entropy": 5.079160737991333, "epoch": 2.0809815950920245, "grad_norm": 0.52734375, "learning_rate": 0.0004792992322187927, "loss": 4.87, "mean_token_accuracy": 0.23461830466985703, "num_tokens": 41361850.0, "step": 2545 }, { "entropy": 5.039652872085571, "epoch": 2.085071574642127, "grad_norm": 0.53515625, "learning_rate": 0.00047916705442645213, "loss": 4.831, "mean_token_accuracy": 0.2346809223294258, "num_tokens": 41442968.0, "step": 2550 }, { "entropy": 5.072763824462891, "epoch": 2.089161554192229, "grad_norm": 0.53125, "learning_rate": 0.0004790344764686219, "loss": 4.8246, "mean_token_accuracy": 0.23943378776311874, "num_tokens": 41524121.0, "step": 2555 }, { "entropy": 5.025501728057861, "epoch": 2.093251533742331, "grad_norm": 0.5546875, "learning_rate": 0.00047890149860515337, "loss": 4.859, "mean_token_accuracy": 0.23201810717582702, "num_tokens": 41605534.0, "step": 2560 }, { "entropy": 5.084275531768799, "epoch": 2.0973415132924336, "grad_norm": 0.51953125, "learning_rate": 0.00047876812109668197, "loss": 4.8554, "mean_token_accuracy": 0.2370379999279976, "num_tokens": 41687217.0, "step": 2565 }, { "entropy": 5.077451372146607, "epoch": 2.101431492842536, "grad_norm": 0.546875, "learning_rate": 0.00047863434420462625, "loss": 4.8462, "mean_token_accuracy": 0.23559490740299224, "num_tokens": 41768865.0, "step": 2570 }, { "entropy": 5.037510919570923, "epoch": 2.105521472392638, "grad_norm": 0.5234375, "learning_rate": 0.0004785001681911878, "loss": 4.8446, "mean_token_accuracy": 0.2362859934568405, "num_tokens": 41849782.0, "step": 2575 }, { "entropy": 5.060849571228028, "epoch": 2.1096114519427402, "grad_norm": 0.56640625, "learning_rate": 0.0004783655933193502, "loss": 4.8618, "mean_token_accuracy": 0.23436325639486313, "num_tokens": 41930962.0, "step": 2580 }, { "entropy": 5.032677745819091, "epoch": 2.1137014314928426, "grad_norm": 0.546875, "learning_rate": 0.00047823061985287905, "loss": 4.8156, "mean_token_accuracy": 0.2397855579853058, "num_tokens": 42012210.0, "step": 2585 }, { "entropy": 5.041714096069336, "epoch": 2.117791411042945, "grad_norm": 0.57421875, "learning_rate": 0.00047809524805632087, "loss": 4.8257, "mean_token_accuracy": 0.23939530402421952, "num_tokens": 42093182.0, "step": 2590 }, { "entropy": 5.042909097671509, "epoch": 2.121881390593047, "grad_norm": 0.58203125, "learning_rate": 0.0004779594781950033, "loss": 4.8416, "mean_token_accuracy": 0.23777866810560228, "num_tokens": 42174316.0, "step": 2595 }, { "entropy": 5.067096090316772, "epoch": 2.1259713701431493, "grad_norm": 0.52734375, "learning_rate": 0.00047782331053503393, "loss": 4.8438, "mean_token_accuracy": 0.23809701204299927, "num_tokens": 42255978.0, "step": 2600 }, { "entropy": 5.050254058837891, "epoch": 2.1300613496932517, "grad_norm": 0.53515625, "learning_rate": 0.00047768674534329994, "loss": 4.8527, "mean_token_accuracy": 0.23612816333770753, "num_tokens": 42337256.0, "step": 2605 }, { "entropy": 5.058381509780884, "epoch": 2.1341513292433536, "grad_norm": 0.51953125, "learning_rate": 0.000477549782887468, "loss": 4.8406, "mean_token_accuracy": 0.23489826917648315, "num_tokens": 42418431.0, "step": 2610 }, { "entropy": 5.045698595046997, "epoch": 2.138241308793456, "grad_norm": 0.5234375, "learning_rate": 0.0004774124234359831, "loss": 4.8904, "mean_token_accuracy": 0.23427049070596695, "num_tokens": 42499627.0, "step": 2615 }, { "entropy": 5.049042510986328, "epoch": 2.1423312883435583, "grad_norm": 0.52734375, "learning_rate": 0.0004772746672580685, "loss": 4.8589, "mean_token_accuracy": 0.23529830425977707, "num_tokens": 42581245.0, "step": 2620 }, { "entropy": 5.069973993301391, "epoch": 2.1464212678936607, "grad_norm": 0.515625, "learning_rate": 0.00047713651462372506, "loss": 4.8801, "mean_token_accuracy": 0.23152053207159043, "num_tokens": 42662950.0, "step": 2625 }, { "entropy": 5.019741678237915, "epoch": 2.1505112474437627, "grad_norm": 0.51171875, "learning_rate": 0.0004769979658037305, "loss": 4.8157, "mean_token_accuracy": 0.23884918242692948, "num_tokens": 42744069.0, "step": 2630 }, { "entropy": 4.9991796016693115, "epoch": 2.154601226993865, "grad_norm": 0.5390625, "learning_rate": 0.0004768590210696393, "loss": 4.8144, "mean_token_accuracy": 0.23908749222755432, "num_tokens": 42825688.0, "step": 2635 }, { "entropy": 5.125963926315308, "epoch": 2.1586912065439674, "grad_norm": 0.52734375, "learning_rate": 0.00047671968069378185, "loss": 4.9056, "mean_token_accuracy": 0.23172895163297652, "num_tokens": 42906531.0, "step": 2640 }, { "entropy": 5.023270797729492, "epoch": 2.1627811860940693, "grad_norm": 0.5234375, "learning_rate": 0.0004765799449492639, "loss": 4.839, "mean_token_accuracy": 0.23729376047849654, "num_tokens": 42988112.0, "step": 2645 }, { "entropy": 5.0261146068573, "epoch": 2.1668711656441717, "grad_norm": 0.5625, "learning_rate": 0.0004764398141099661, "loss": 4.8315, "mean_token_accuracy": 0.23749377876520156, "num_tokens": 43069032.0, "step": 2650 }, { "entropy": 5.098901414871216, "epoch": 2.170961145194274, "grad_norm": 0.52734375, "learning_rate": 0.0004762992884505438, "loss": 4.8685, "mean_token_accuracy": 0.23616028279066087, "num_tokens": 43150625.0, "step": 2655 }, { "entropy": 4.996641826629639, "epoch": 2.1750511247443765, "grad_norm": 0.53125, "learning_rate": 0.00047615836824642567, "loss": 4.8109, "mean_token_accuracy": 0.23855308294296265, "num_tokens": 43232196.0, "step": 2660 }, { "entropy": 5.045117044448853, "epoch": 2.1791411042944784, "grad_norm": 0.52734375, "learning_rate": 0.00047601705377381414, "loss": 4.8348, "mean_token_accuracy": 0.2365160033106804, "num_tokens": 43313726.0, "step": 2665 }, { "entropy": 4.970278596878051, "epoch": 2.1832310838445808, "grad_norm": 0.5234375, "learning_rate": 0.0004758753453096843, "loss": 4.8005, "mean_token_accuracy": 0.2412037208676338, "num_tokens": 43395075.0, "step": 2670 }, { "entropy": 5.042024612426758, "epoch": 2.187321063394683, "grad_norm": 0.5, "learning_rate": 0.00047573324313178325, "loss": 4.8444, "mean_token_accuracy": 0.23562384992837906, "num_tokens": 43475949.0, "step": 2675 }, { "entropy": 5.049530363082885, "epoch": 2.191411042944785, "grad_norm": 0.53515625, "learning_rate": 0.00047559074751862997, "loss": 4.855, "mean_token_accuracy": 0.23930780291557313, "num_tokens": 43557611.0, "step": 2680 }, { "entropy": 5.00416669845581, "epoch": 2.1955010224948874, "grad_norm": 0.53515625, "learning_rate": 0.0004754478587495145, "loss": 4.8077, "mean_token_accuracy": 0.23757991939783096, "num_tokens": 43639239.0, "step": 2685 }, { "entropy": 5.072630405426025, "epoch": 2.19959100204499, "grad_norm": 0.55078125, "learning_rate": 0.00047530457710449756, "loss": 4.8478, "mean_token_accuracy": 0.23567817360162735, "num_tokens": 43720608.0, "step": 2690 }, { "entropy": 5.057205247879028, "epoch": 2.203680981595092, "grad_norm": 0.53515625, "learning_rate": 0.00047516090286440967, "loss": 4.8685, "mean_token_accuracy": 0.23655719608068465, "num_tokens": 43802224.0, "step": 2695 }, { "entropy": 5.035128355026245, "epoch": 2.207770961145194, "grad_norm": 0.51953125, "learning_rate": 0.00047501683631085106, "loss": 4.8328, "mean_token_accuracy": 0.23788557797670365, "num_tokens": 43883400.0, "step": 2700 }, { "entropy": 5.024647665023804, "epoch": 2.2118609406952965, "grad_norm": 0.5234375, "learning_rate": 0.0004748723777261908, "loss": 4.8089, "mean_token_accuracy": 0.23928927183151244, "num_tokens": 43965068.0, "step": 2705 }, { "entropy": 4.972833681106567, "epoch": 2.215950920245399, "grad_norm": 0.5234375, "learning_rate": 0.00047472752739356636, "loss": 4.7783, "mean_token_accuracy": 0.23996302634477615, "num_tokens": 44046676.0, "step": 2710 }, { "entropy": 5.08757791519165, "epoch": 2.2200408997955012, "grad_norm": 0.515625, "learning_rate": 0.00047458228559688287, "loss": 4.8814, "mean_token_accuracy": 0.23545849472284316, "num_tokens": 44128234.0, "step": 2715 }, { "entropy": 5.004882764816284, "epoch": 2.224130879345603, "grad_norm": 0.52734375, "learning_rate": 0.00047443665262081314, "loss": 4.8573, "mean_token_accuracy": 0.23648130744695664, "num_tokens": 44209856.0, "step": 2720 }, { "entropy": 5.032129573822021, "epoch": 2.2282208588957055, "grad_norm": 0.52734375, "learning_rate": 0.00047429062875079626, "loss": 4.8592, "mean_token_accuracy": 0.23640067279338836, "num_tokens": 44291300.0, "step": 2725 }, { "entropy": 5.023078346252442, "epoch": 2.232310838445808, "grad_norm": 0.5078125, "learning_rate": 0.0004741442142730376, "loss": 4.8597, "mean_token_accuracy": 0.23393862843513488, "num_tokens": 44372947.0, "step": 2730 }, { "entropy": 5.014073514938355, "epoch": 2.23640081799591, "grad_norm": 0.53515625, "learning_rate": 0.0004739974094745082, "loss": 4.8202, "mean_token_accuracy": 0.23982056379318237, "num_tokens": 44454488.0, "step": 2735 }, { "entropy": 4.985676670074463, "epoch": 2.240490797546012, "grad_norm": 0.51953125, "learning_rate": 0.00047385021464294415, "loss": 4.8005, "mean_token_accuracy": 0.2414940506219864, "num_tokens": 44535839.0, "step": 2740 }, { "entropy": 5.020768880844116, "epoch": 2.2445807770961146, "grad_norm": 0.52734375, "learning_rate": 0.00047370263006684576, "loss": 4.8202, "mean_token_accuracy": 0.23832704573869706, "num_tokens": 44616996.0, "step": 2745 }, { "entropy": 5.035280084609985, "epoch": 2.248670756646217, "grad_norm": 0.5078125, "learning_rate": 0.0004735546560354775, "loss": 4.8554, "mean_token_accuracy": 0.23616455644369125, "num_tokens": 44698351.0, "step": 2750 }, { "entropy": 4.980247116088867, "epoch": 2.252760736196319, "grad_norm": 0.5234375, "learning_rate": 0.0004734062928388671, "loss": 4.7991, "mean_token_accuracy": 0.24156965166330338, "num_tokens": 44779663.0, "step": 2755 }, { "entropy": 5.021684646606445, "epoch": 2.2568507157464213, "grad_norm": 0.51953125, "learning_rate": 0.000473257540767805, "loss": 4.817, "mean_token_accuracy": 0.23756028860807418, "num_tokens": 44861184.0, "step": 2760 }, { "entropy": 4.986014080047608, "epoch": 2.2609406952965236, "grad_norm": 0.51171875, "learning_rate": 0.0004731084001138437, "loss": 4.8172, "mean_token_accuracy": 0.2385167196393013, "num_tokens": 44942868.0, "step": 2765 }, { "entropy": 5.006426382064819, "epoch": 2.265030674846626, "grad_norm": 0.53125, "learning_rate": 0.0004729588711692976, "loss": 4.8079, "mean_token_accuracy": 0.2389816462993622, "num_tokens": 45023944.0, "step": 2770 }, { "entropy": 4.962817907333374, "epoch": 2.269120654396728, "grad_norm": 0.5390625, "learning_rate": 0.00047280895422724205, "loss": 4.783, "mean_token_accuracy": 0.2405168741941452, "num_tokens": 45105493.0, "step": 2775 }, { "entropy": 5.010297536849976, "epoch": 2.2732106339468303, "grad_norm": 0.51953125, "learning_rate": 0.00047265864958151286, "loss": 4.84, "mean_token_accuracy": 0.2374628499150276, "num_tokens": 45186775.0, "step": 2780 }, { "entropy": 4.9936075687408445, "epoch": 2.2773006134969327, "grad_norm": 0.5234375, "learning_rate": 0.0004725079575267055, "loss": 4.7956, "mean_token_accuracy": 0.23815418481826783, "num_tokens": 45267675.0, "step": 2785 }, { "entropy": 4.999183607101441, "epoch": 2.2813905930470346, "grad_norm": 0.498046875, "learning_rate": 0.0004723568783581751, "loss": 4.8209, "mean_token_accuracy": 0.2396041676402092, "num_tokens": 45348942.0, "step": 2790 }, { "entropy": 5.004906177520752, "epoch": 2.285480572597137, "grad_norm": 0.5078125, "learning_rate": 0.00047220541237203545, "loss": 4.8113, "mean_token_accuracy": 0.23775430917739868, "num_tokens": 45430509.0, "step": 2795 }, { "entropy": 5.0051342964172365, "epoch": 2.2895705521472394, "grad_norm": 0.51171875, "learning_rate": 0.0004720535598651584, "loss": 4.8564, "mean_token_accuracy": 0.2341430053114891, "num_tokens": 45511909.0, "step": 2800 }, { "entropy": 4.977736902236939, "epoch": 2.2936605316973413, "grad_norm": 0.52734375, "learning_rate": 0.0004719013211351733, "loss": 4.786, "mean_token_accuracy": 0.24154940843582154, "num_tokens": 45593354.0, "step": 2805 }, { "entropy": 5.057413816452026, "epoch": 2.2977505112474437, "grad_norm": 0.5078125, "learning_rate": 0.00047174869648046686, "loss": 4.8356, "mean_token_accuracy": 0.24007793962955476, "num_tokens": 45674546.0, "step": 2810 }, { "entropy": 5.0005053043365475, "epoch": 2.301840490797546, "grad_norm": 0.5234375, "learning_rate": 0.0004715956862001818, "loss": 4.8381, "mean_token_accuracy": 0.23803626149892806, "num_tokens": 45755237.0, "step": 2815 }, { "entropy": 4.988029050827026, "epoch": 2.3059304703476484, "grad_norm": 0.53125, "learning_rate": 0.00047144229059421685, "loss": 4.813, "mean_token_accuracy": 0.2401291847229004, "num_tokens": 45836869.0, "step": 2820 }, { "entropy": 5.080396366119385, "epoch": 2.3100204498977503, "grad_norm": 0.5625, "learning_rate": 0.00047128850996322594, "loss": 4.8511, "mean_token_accuracy": 0.2373419538140297, "num_tokens": 45918545.0, "step": 2825 }, { "entropy": 4.985589551925659, "epoch": 2.3141104294478527, "grad_norm": 0.546875, "learning_rate": 0.0004711343446086176, "loss": 4.8632, "mean_token_accuracy": 0.2379470333456993, "num_tokens": 46000109.0, "step": 2830 }, { "entropy": 5.038642692565918, "epoch": 2.318200408997955, "grad_norm": 0.51171875, "learning_rate": 0.0004709797948325545, "loss": 4.8402, "mean_token_accuracy": 0.23703430145978927, "num_tokens": 46081580.0, "step": 2835 }, { "entropy": 5.018342351913452, "epoch": 2.3222903885480575, "grad_norm": 0.51171875, "learning_rate": 0.00047082486093795277, "loss": 4.8141, "mean_token_accuracy": 0.2407286897301674, "num_tokens": 46163170.0, "step": 2840 }, { "entropy": 4.977063322067261, "epoch": 2.3263803680981594, "grad_norm": 0.51953125, "learning_rate": 0.00047066954322848135, "loss": 4.7828, "mean_token_accuracy": 0.24125634580850602, "num_tokens": 46244281.0, "step": 2845 }, { "entropy": 5.005881071090698, "epoch": 2.3304703476482618, "grad_norm": 0.52734375, "learning_rate": 0.00047051384200856155, "loss": 4.8195, "mean_token_accuracy": 0.24058586806058885, "num_tokens": 46325996.0, "step": 2850 }, { "entropy": 4.988586187362671, "epoch": 2.334560327198364, "grad_norm": 0.51953125, "learning_rate": 0.00047035775758336607, "loss": 4.7979, "mean_token_accuracy": 0.23823931515216829, "num_tokens": 46407298.0, "step": 2855 }, { "entropy": 4.954330635070801, "epoch": 2.338650306748466, "grad_norm": 0.51953125, "learning_rate": 0.00047020129025881916, "loss": 4.7947, "mean_token_accuracy": 0.2432621404528618, "num_tokens": 46488692.0, "step": 2860 }, { "entropy": 5.001978206634521, "epoch": 2.3427402862985685, "grad_norm": 0.515625, "learning_rate": 0.0004700444403415951, "loss": 4.8296, "mean_token_accuracy": 0.23775594234466552, "num_tokens": 46569660.0, "step": 2865 }, { "entropy": 5.007827568054199, "epoch": 2.346830265848671, "grad_norm": 0.50390625, "learning_rate": 0.00046988720813911835, "loss": 4.8172, "mean_token_accuracy": 0.23724746704101562, "num_tokens": 46650717.0, "step": 2870 }, { "entropy": 5.0017200946807865, "epoch": 2.3509202453987728, "grad_norm": 0.54296875, "learning_rate": 0.0004697295939595624, "loss": 4.8164, "mean_token_accuracy": 0.23755236566066742, "num_tokens": 46731857.0, "step": 2875 }, { "entropy": 4.963125133514405, "epoch": 2.355010224948875, "grad_norm": 0.515625, "learning_rate": 0.0004695715981118497, "loss": 4.7906, "mean_token_accuracy": 0.24304623603820802, "num_tokens": 46812819.0, "step": 2880 }, { "entropy": 4.976752138137817, "epoch": 2.3591002044989775, "grad_norm": 0.51171875, "learning_rate": 0.0004694132209056506, "loss": 4.8264, "mean_token_accuracy": 0.24056561291217804, "num_tokens": 46894283.0, "step": 2885 }, { "entropy": 5.035718727111816, "epoch": 2.36319018404908, "grad_norm": 0.515625, "learning_rate": 0.0004692544626513828, "loss": 4.8353, "mean_token_accuracy": 0.2396184742450714, "num_tokens": 46975974.0, "step": 2890 }, { "entropy": 4.935823440551758, "epoch": 2.3672801635991823, "grad_norm": 0.515625, "learning_rate": 0.00046909532366021115, "loss": 4.8073, "mean_token_accuracy": 0.24061134457588196, "num_tokens": 47057478.0, "step": 2895 }, { "entropy": 5.057424592971802, "epoch": 2.371370143149284, "grad_norm": 0.51953125, "learning_rate": 0.00046893580424404644, "loss": 4.8815, "mean_token_accuracy": 0.2351416289806366, "num_tokens": 47138643.0, "step": 2900 }, { "entropy": 5.03563961982727, "epoch": 2.3754601226993866, "grad_norm": 0.5078125, "learning_rate": 0.00046877590471554515, "loss": 4.8474, "mean_token_accuracy": 0.23867853432893754, "num_tokens": 47218821.0, "step": 2905 }, { "entropy": 4.971715831756592, "epoch": 2.379550102249489, "grad_norm": 0.55078125, "learning_rate": 0.00046861562538810895, "loss": 4.7758, "mean_token_accuracy": 0.24315285384655, "num_tokens": 47300491.0, "step": 2910 }, { "entropy": 4.957176208496094, "epoch": 2.383640081799591, "grad_norm": 0.51953125, "learning_rate": 0.0004684549665758838, "loss": 4.7926, "mean_token_accuracy": 0.2414930135011673, "num_tokens": 47381772.0, "step": 2915 }, { "entropy": 5.002673244476318, "epoch": 2.3877300613496932, "grad_norm": 0.50390625, "learning_rate": 0.0004682939285937594, "loss": 4.8269, "mean_token_accuracy": 0.24218309968709945, "num_tokens": 47463330.0, "step": 2920 }, { "entropy": 4.946222352981567, "epoch": 2.3918200408997956, "grad_norm": 0.52734375, "learning_rate": 0.00046813251175736875, "loss": 4.7785, "mean_token_accuracy": 0.24289505779743195, "num_tokens": 47544795.0, "step": 2925 }, { "entropy": 5.01896185874939, "epoch": 2.3959100204498975, "grad_norm": 0.50390625, "learning_rate": 0.0004679707163830872, "loss": 4.8379, "mean_token_accuracy": 0.234355828166008, "num_tokens": 47625658.0, "step": 2930 }, { "entropy": 4.966737461090088, "epoch": 2.4, "grad_norm": 0.51171875, "learning_rate": 0.00046780854278803224, "loss": 4.8066, "mean_token_accuracy": 0.24003686308860778, "num_tokens": 47707318.0, "step": 2935 }, { "entropy": 5.028832864761353, "epoch": 2.4040899795501023, "grad_norm": 0.50390625, "learning_rate": 0.0004676459912900625, "loss": 4.8245, "mean_token_accuracy": 0.24042247980833054, "num_tokens": 47788406.0, "step": 2940 }, { "entropy": 4.990110301971436, "epoch": 2.4081799591002047, "grad_norm": 0.51171875, "learning_rate": 0.00046748306220777743, "loss": 4.8149, "mean_token_accuracy": 0.24144220501184463, "num_tokens": 47870018.0, "step": 2945 }, { "entropy": 5.040558671951294, "epoch": 2.4122699386503066, "grad_norm": 0.5234375, "learning_rate": 0.0004673197558605165, "loss": 4.878, "mean_token_accuracy": 0.2372189536690712, "num_tokens": 47951519.0, "step": 2950 }, { "entropy": 5.013999938964844, "epoch": 2.416359918200409, "grad_norm": 0.5390625, "learning_rate": 0.0004671560725683586, "loss": 4.8303, "mean_token_accuracy": 0.2398605987429619, "num_tokens": 48032168.0, "step": 2955 }, { "entropy": 4.937393569946289, "epoch": 2.4204498977505113, "grad_norm": 0.5078125, "learning_rate": 0.0004669920126521213, "loss": 4.7484, "mean_token_accuracy": 0.24363629519939423, "num_tokens": 48113394.0, "step": 2960 }, { "entropy": 4.9911376476287845, "epoch": 2.4245398773006137, "grad_norm": 0.5234375, "learning_rate": 0.00046682757643336074, "loss": 4.8293, "mean_token_accuracy": 0.23650192171335221, "num_tokens": 48194765.0, "step": 2965 }, { "entropy": 4.970001268386841, "epoch": 2.4286298568507156, "grad_norm": 0.494140625, "learning_rate": 0.0004666627642343701, "loss": 4.7711, "mean_token_accuracy": 0.24166625142097473, "num_tokens": 48275779.0, "step": 2970 }, { "entropy": 4.909131956100464, "epoch": 2.432719836400818, "grad_norm": 0.51953125, "learning_rate": 0.00046649757637817985, "loss": 4.7414, "mean_token_accuracy": 0.24418042451143265, "num_tokens": 48356504.0, "step": 2975 }, { "entropy": 5.032081937789917, "epoch": 2.4368098159509204, "grad_norm": 0.53125, "learning_rate": 0.0004663320131885565, "loss": 4.8359, "mean_token_accuracy": 0.23731423318386077, "num_tokens": 48438039.0, "step": 2980 }, { "entropy": 4.9816405296325685, "epoch": 2.4408997955010223, "grad_norm": 0.5078125, "learning_rate": 0.00046616607499000255, "loss": 4.7925, "mean_token_accuracy": 0.24088211208581925, "num_tokens": 48519713.0, "step": 2985 }, { "entropy": 4.957751274108887, "epoch": 2.4449897750511247, "grad_norm": 0.5078125, "learning_rate": 0.0004659997621077551, "loss": 4.8015, "mean_token_accuracy": 0.23747823089361192, "num_tokens": 48600662.0, "step": 2990 }, { "entropy": 5.003507614135742, "epoch": 2.449079754601227, "grad_norm": 0.52734375, "learning_rate": 0.00046583307486778593, "loss": 4.8315, "mean_token_accuracy": 0.24149917662143708, "num_tokens": 48681819.0, "step": 2995 }, { "entropy": 4.995101642608643, "epoch": 2.453169734151329, "grad_norm": 0.51171875, "learning_rate": 0.0004656660135968004, "loss": 4.7899, "mean_token_accuracy": 0.24259499460458755, "num_tokens": 48763430.0, "step": 3000 }, { "epoch": 2.453169734151329, "eval_entropy": 4.824206147875104, "eval_loss": 4.888221740722656, "eval_mean_token_accuracy": 0.23986414189964647, "eval_num_tokens": 48763430.0, "eval_runtime": 66.6461, "eval_samples_per_second": 62.074, "eval_steps_per_second": 7.772, "step": 3000 }, { "entropy": 4.958430051803589, "epoch": 2.4572597137014314, "grad_norm": 0.5078125, "learning_rate": 0.00046549857862223694, "loss": 4.7891, "mean_token_accuracy": 0.23962418138980865, "num_tokens": 48843965.0, "step": 3005 }, { "entropy": 4.956654453277588, "epoch": 2.4613496932515337, "grad_norm": 0.51953125, "learning_rate": 0.00046533077027226667, "loss": 4.7642, "mean_token_accuracy": 0.241588594019413, "num_tokens": 48925391.0, "step": 3010 }, { "entropy": 4.970054292678833, "epoch": 2.465439672801636, "grad_norm": 0.5390625, "learning_rate": 0.0004651625888757923, "loss": 4.8226, "mean_token_accuracy": 0.23956118673086166, "num_tokens": 49006973.0, "step": 3015 }, { "entropy": 4.96092267036438, "epoch": 2.469529652351738, "grad_norm": 0.51953125, "learning_rate": 0.00046499403476244774, "loss": 4.8062, "mean_token_accuracy": 0.24216772466897965, "num_tokens": 49088409.0, "step": 3020 }, { "entropy": 4.961851739883423, "epoch": 2.4736196319018404, "grad_norm": 0.515625, "learning_rate": 0.0004648251082625975, "loss": 4.7905, "mean_token_accuracy": 0.23836839646101, "num_tokens": 49169952.0, "step": 3025 }, { "entropy": 4.943976497650146, "epoch": 2.477709611451943, "grad_norm": 0.50390625, "learning_rate": 0.0004646558097073358, "loss": 4.7998, "mean_token_accuracy": 0.23892782479524613, "num_tokens": 49251506.0, "step": 3030 }, { "entropy": 4.940610027313232, "epoch": 2.481799591002045, "grad_norm": 0.5078125, "learning_rate": 0.00046448613942848635, "loss": 4.7726, "mean_token_accuracy": 0.24038388282060624, "num_tokens": 49332179.0, "step": 3035 }, { "entropy": 4.962407875061035, "epoch": 2.485889570552147, "grad_norm": 0.53125, "learning_rate": 0.0004643160977586013, "loss": 4.7963, "mean_token_accuracy": 0.24074705243110656, "num_tokens": 49413789.0, "step": 3040 }, { "entropy": 4.971686172485351, "epoch": 2.4899795501022495, "grad_norm": 0.50390625, "learning_rate": 0.0004641456850309605, "loss": 4.7962, "mean_token_accuracy": 0.24148085713386536, "num_tokens": 49495281.0, "step": 3045 }, { "entropy": 4.926886701583863, "epoch": 2.494069529652352, "grad_norm": 0.51953125, "learning_rate": 0.0004639749015795715, "loss": 4.7565, "mean_token_accuracy": 0.2452280268073082, "num_tokens": 49576707.0, "step": 3050 }, { "entropy": 4.940164184570312, "epoch": 2.4981595092024538, "grad_norm": 0.49609375, "learning_rate": 0.00046380374773916815, "loss": 4.7843, "mean_token_accuracy": 0.2416828006505966, "num_tokens": 49657749.0, "step": 3055 }, { "entropy": 4.956660604476928, "epoch": 2.502249488752556, "grad_norm": 0.52734375, "learning_rate": 0.0004636322238452104, "loss": 4.7796, "mean_token_accuracy": 0.24415811598300935, "num_tokens": 49738903.0, "step": 3060 }, { "entropy": 4.92326283454895, "epoch": 2.5063394683026585, "grad_norm": 0.484375, "learning_rate": 0.0004634603302338834, "loss": 4.7591, "mean_token_accuracy": 0.24299813359975814, "num_tokens": 49820124.0, "step": 3065 }, { "entropy": 4.950853109359741, "epoch": 2.510429447852761, "grad_norm": 0.515625, "learning_rate": 0.000463288067242097, "loss": 4.7501, "mean_token_accuracy": 0.24229330867528914, "num_tokens": 49901184.0, "step": 3070 }, { "entropy": 4.9711895942687985, "epoch": 2.514519427402863, "grad_norm": 0.51953125, "learning_rate": 0.00046311543520748507, "loss": 4.769, "mean_token_accuracy": 0.24455423802137374, "num_tokens": 49981940.0, "step": 3075 }, { "entropy": 4.90674090385437, "epoch": 2.518609406952965, "grad_norm": 0.51171875, "learning_rate": 0.0004629424344684047, "loss": 4.7671, "mean_token_accuracy": 0.2435960590839386, "num_tokens": 50063116.0, "step": 3080 }, { "entropy": 5.018128156661987, "epoch": 2.5226993865030676, "grad_norm": 0.5078125, "learning_rate": 0.0004627690653639358, "loss": 4.8314, "mean_token_accuracy": 0.24213892966508865, "num_tokens": 50144795.0, "step": 3085 }, { "entropy": 4.9256250858306885, "epoch": 2.52678936605317, "grad_norm": 0.52734375, "learning_rate": 0.00046259532823388007, "loss": 4.7526, "mean_token_accuracy": 0.24392250329256057, "num_tokens": 50226181.0, "step": 3090 }, { "entropy": 4.98425669670105, "epoch": 2.530879345603272, "grad_norm": 0.5078125, "learning_rate": 0.0004624212234187607, "loss": 4.8185, "mean_token_accuracy": 0.24152301400899887, "num_tokens": 50307381.0, "step": 3095 }, { "entropy": 4.989864540100098, "epoch": 2.5349693251533743, "grad_norm": 0.50390625, "learning_rate": 0.0004622467512598215, "loss": 4.7953, "mean_token_accuracy": 0.24576946645975112, "num_tokens": 50389016.0, "step": 3100 }, { "entropy": 4.949513864517212, "epoch": 2.5390593047034766, "grad_norm": 0.51171875, "learning_rate": 0.0004620719120990261, "loss": 4.7827, "mean_token_accuracy": 0.24587612897157668, "num_tokens": 50470006.0, "step": 3105 }, { "entropy": 4.939216899871826, "epoch": 2.5431492842535786, "grad_norm": 0.498046875, "learning_rate": 0.0004618967062790577, "loss": 4.7937, "mean_token_accuracy": 0.2423356592655182, "num_tokens": 50551648.0, "step": 3110 }, { "entropy": 4.968730163574219, "epoch": 2.547239263803681, "grad_norm": 0.515625, "learning_rate": 0.00046172113414331804, "loss": 4.7788, "mean_token_accuracy": 0.2447182059288025, "num_tokens": 50632416.0, "step": 3115 }, { "entropy": 4.937106418609619, "epoch": 2.5513292433537833, "grad_norm": 0.515625, "learning_rate": 0.0004615451960359269, "loss": 4.8131, "mean_token_accuracy": 0.23890747129917145, "num_tokens": 50714065.0, "step": 3120 }, { "entropy": 4.976661205291748, "epoch": 2.5554192229038852, "grad_norm": 0.53125, "learning_rate": 0.00046136889230172126, "loss": 4.7972, "mean_token_accuracy": 0.24100573658943175, "num_tokens": 50795553.0, "step": 3125 }, { "entropy": 4.986274147033692, "epoch": 2.5595092024539876, "grad_norm": 0.53515625, "learning_rate": 0.00046119222328625464, "loss": 4.7999, "mean_token_accuracy": 0.24317229986190797, "num_tokens": 50876829.0, "step": 3130 }, { "entropy": 4.90229868888855, "epoch": 2.56359918200409, "grad_norm": 0.51171875, "learning_rate": 0.0004610151893357968, "loss": 4.7699, "mean_token_accuracy": 0.24466679394245147, "num_tokens": 50958104.0, "step": 3135 }, { "entropy": 4.960571575164795, "epoch": 2.5676891615541924, "grad_norm": 0.52734375, "learning_rate": 0.0004608377907973326, "loss": 4.7767, "mean_token_accuracy": 0.24391674250364304, "num_tokens": 51039045.0, "step": 3140 }, { "entropy": 4.944932222366333, "epoch": 2.5717791411042947, "grad_norm": 0.5234375, "learning_rate": 0.00046066002801856126, "loss": 4.7908, "mean_token_accuracy": 0.24250412732362747, "num_tokens": 51120621.0, "step": 3145 }, { "entropy": 4.997454690933227, "epoch": 2.5758691206543967, "grad_norm": 0.498046875, "learning_rate": 0.0004604819013478965, "loss": 4.8396, "mean_token_accuracy": 0.241471067070961, "num_tokens": 51202206.0, "step": 3150 }, { "entropy": 5.009496736526489, "epoch": 2.579959100204499, "grad_norm": 0.5390625, "learning_rate": 0.0004603034111344646, "loss": 4.8304, "mean_token_accuracy": 0.23892204761505126, "num_tokens": 51282817.0, "step": 3155 }, { "entropy": 4.988063716888428, "epoch": 2.5840490797546014, "grad_norm": 0.52734375, "learning_rate": 0.00046012455772810487, "loss": 4.8141, "mean_token_accuracy": 0.2395667925477028, "num_tokens": 51363867.0, "step": 3160 }, { "entropy": 4.975194454193115, "epoch": 2.5881390593047033, "grad_norm": 0.515625, "learning_rate": 0.0004599453414793682, "loss": 4.7803, "mean_token_accuracy": 0.24350594133138656, "num_tokens": 51445004.0, "step": 3165 }, { "entropy": 4.943451023101806, "epoch": 2.5922290388548057, "grad_norm": 0.5078125, "learning_rate": 0.0004597657627395168, "loss": 4.8126, "mean_token_accuracy": 0.24055346399545668, "num_tokens": 51526156.0, "step": 3170 }, { "entropy": 4.937479782104492, "epoch": 2.596319018404908, "grad_norm": 0.5078125, "learning_rate": 0.0004595858218605232, "loss": 4.7496, "mean_token_accuracy": 0.2447401165962219, "num_tokens": 51607815.0, "step": 3175 }, { "entropy": 4.958913135528564, "epoch": 2.60040899795501, "grad_norm": 0.498046875, "learning_rate": 0.0004594055191950699, "loss": 4.7767, "mean_token_accuracy": 0.2425516813993454, "num_tokens": 51688563.0, "step": 3180 }, { "entropy": 4.911005544662475, "epoch": 2.6044989775051124, "grad_norm": 0.54296875, "learning_rate": 0.0004592248550965484, "loss": 4.7515, "mean_token_accuracy": 0.24578267931938172, "num_tokens": 51770151.0, "step": 3185 }, { "entropy": 4.951757478713989, "epoch": 2.6085889570552148, "grad_norm": 0.5, "learning_rate": 0.00045904382991905857, "loss": 4.7568, "mean_token_accuracy": 0.24362249076366424, "num_tokens": 51851380.0, "step": 3190 }, { "entropy": 4.912914371490478, "epoch": 2.612678936605317, "grad_norm": 0.51171875, "learning_rate": 0.0004588624440174081, "loss": 4.7794, "mean_token_accuracy": 0.24376973658800125, "num_tokens": 51932418.0, "step": 3195 }, { "entropy": 4.966513824462891, "epoch": 2.616768916155419, "grad_norm": 0.51171875, "learning_rate": 0.0004586806977471117, "loss": 4.801, "mean_token_accuracy": 0.24054027944803238, "num_tokens": 52012876.0, "step": 3200 }, { "entropy": 4.94195327758789, "epoch": 2.6208588957055214, "grad_norm": 0.50390625, "learning_rate": 0.00045849859146439016, "loss": 4.7895, "mean_token_accuracy": 0.23972345441579818, "num_tokens": 52094116.0, "step": 3205 }, { "entropy": 4.95401291847229, "epoch": 2.624948875255624, "grad_norm": 0.53125, "learning_rate": 0.0004583161255261702, "loss": 4.7928, "mean_token_accuracy": 0.24293293803930283, "num_tokens": 52174555.0, "step": 3210 }, { "entropy": 4.97545051574707, "epoch": 2.629038854805726, "grad_norm": 0.51171875, "learning_rate": 0.0004581333002900832, "loss": 4.8035, "mean_token_accuracy": 0.2448437601327896, "num_tokens": 52255621.0, "step": 3215 }, { "entropy": 4.933709955215454, "epoch": 2.633128834355828, "grad_norm": 0.50390625, "learning_rate": 0.0004579501161144651, "loss": 4.7847, "mean_token_accuracy": 0.24162852466106416, "num_tokens": 52336931.0, "step": 3220 }, { "entropy": 4.9425678730010985, "epoch": 2.6372188139059305, "grad_norm": 0.51171875, "learning_rate": 0.0004577665733583549, "loss": 4.7436, "mean_token_accuracy": 0.24749198108911513, "num_tokens": 52418260.0, "step": 3225 }, { "entropy": 4.957076644897461, "epoch": 2.641308793456033, "grad_norm": 0.51171875, "learning_rate": 0.0004575826723814949, "loss": 4.7853, "mean_token_accuracy": 0.24277310073375702, "num_tokens": 52499124.0, "step": 3230 }, { "entropy": 4.947375726699829, "epoch": 2.645398773006135, "grad_norm": 0.515625, "learning_rate": 0.00045739841354432907, "loss": 4.7806, "mean_token_accuracy": 0.24149806350469588, "num_tokens": 52580638.0, "step": 3235 }, { "entropy": 4.884844017028809, "epoch": 2.649488752556237, "grad_norm": 0.515625, "learning_rate": 0.0004572137972080031, "loss": 4.7262, "mean_token_accuracy": 0.24551659524440766, "num_tokens": 52661098.0, "step": 3240 }, { "entropy": 4.934833669662476, "epoch": 2.6535787321063395, "grad_norm": 0.671875, "learning_rate": 0.00045702882373436317, "loss": 4.7491, "mean_token_accuracy": 0.24526319205760955, "num_tokens": 52742762.0, "step": 3245 }, { "entropy": 4.897374248504638, "epoch": 2.6576687116564415, "grad_norm": 0.51953125, "learning_rate": 0.0004568434934859557, "loss": 4.7479, "mean_token_accuracy": 0.2471170336008072, "num_tokens": 52823761.0, "step": 3250 }, { "entropy": 4.94641900062561, "epoch": 2.661758691206544, "grad_norm": 0.494140625, "learning_rate": 0.00045665780682602593, "loss": 4.7738, "mean_token_accuracy": 0.24541191458702089, "num_tokens": 52905349.0, "step": 3255 }, { "entropy": 4.971123218536377, "epoch": 2.665848670756646, "grad_norm": 0.5078125, "learning_rate": 0.00045647176411851826, "loss": 4.8122, "mean_token_accuracy": 0.239681214094162, "num_tokens": 52985608.0, "step": 3260 }, { "entropy": 4.904706192016602, "epoch": 2.6699386503067486, "grad_norm": 0.51953125, "learning_rate": 0.00045628536572807447, "loss": 4.7842, "mean_token_accuracy": 0.24332953691482545, "num_tokens": 53067187.0, "step": 3265 }, { "entropy": 4.992145252227783, "epoch": 2.674028629856851, "grad_norm": 0.5078125, "learning_rate": 0.0004560986120200336, "loss": 4.7832, "mean_token_accuracy": 0.24261016100645066, "num_tokens": 53148895.0, "step": 3270 }, { "entropy": 4.906949663162232, "epoch": 2.678118609406953, "grad_norm": 0.498046875, "learning_rate": 0.0004559115033604313, "loss": 4.7566, "mean_token_accuracy": 0.24651971459388733, "num_tokens": 53230038.0, "step": 3275 }, { "entropy": 4.9824175357818605, "epoch": 2.6822085889570553, "grad_norm": 0.546875, "learning_rate": 0.00045572404011599865, "loss": 4.7948, "mean_token_accuracy": 0.2416501298546791, "num_tokens": 53311313.0, "step": 3280 }, { "entropy": 4.914922380447388, "epoch": 2.6862985685071576, "grad_norm": 0.5, "learning_rate": 0.0004555362226541618, "loss": 4.7256, "mean_token_accuracy": 0.2469981774687767, "num_tokens": 53393021.0, "step": 3285 }, { "entropy": 4.95537633895874, "epoch": 2.6903885480572596, "grad_norm": 0.515625, "learning_rate": 0.00045534805134304126, "loss": 4.7842, "mean_token_accuracy": 0.24201349169015884, "num_tokens": 53474674.0, "step": 3290 }, { "entropy": 4.930451154708862, "epoch": 2.694478527607362, "grad_norm": 0.490234375, "learning_rate": 0.0004551595265514511, "loss": 4.7732, "mean_token_accuracy": 0.2421216830611229, "num_tokens": 53556112.0, "step": 3295 }, { "entropy": 4.928124046325683, "epoch": 2.6985685071574643, "grad_norm": 0.498046875, "learning_rate": 0.00045497064864889796, "loss": 4.7425, "mean_token_accuracy": 0.24643007069826126, "num_tokens": 53637671.0, "step": 3300 }, { "entropy": 4.89126877784729, "epoch": 2.7026584867075663, "grad_norm": 0.50390625, "learning_rate": 0.0004547814180055808, "loss": 4.7693, "mean_token_accuracy": 0.2466226577758789, "num_tokens": 53718833.0, "step": 3305 }, { "entropy": 4.953768682479859, "epoch": 2.7067484662576686, "grad_norm": 0.51171875, "learning_rate": 0.00045459183499238997, "loss": 4.7696, "mean_token_accuracy": 0.23995054066181182, "num_tokens": 53800444.0, "step": 3310 }, { "entropy": 4.954036808013916, "epoch": 2.710838445807771, "grad_norm": 0.49609375, "learning_rate": 0.00045440189998090626, "loss": 4.8059, "mean_token_accuracy": 0.24491851329803466, "num_tokens": 53881904.0, "step": 3315 }, { "entropy": 4.93680362701416, "epoch": 2.7149284253578734, "grad_norm": 0.51953125, "learning_rate": 0.0004542116133434004, "loss": 4.7667, "mean_token_accuracy": 0.24324952512979509, "num_tokens": 53963662.0, "step": 3320 }, { "entropy": 4.941397428512573, "epoch": 2.7190184049079753, "grad_norm": 0.490234375, "learning_rate": 0.0004540209754528328, "loss": 4.7659, "mean_token_accuracy": 0.24453075379133224, "num_tokens": 54045201.0, "step": 3325 }, { "entropy": 4.8985496997833256, "epoch": 2.7231083844580777, "grad_norm": 0.490234375, "learning_rate": 0.0004538299866828514, "loss": 4.729, "mean_token_accuracy": 0.24560168832540513, "num_tokens": 54126787.0, "step": 3330 }, { "entropy": 4.897676134109497, "epoch": 2.72719836400818, "grad_norm": 0.5859375, "learning_rate": 0.0004536386474077927, "loss": 4.7409, "mean_token_accuracy": 0.24957384020090104, "num_tokens": 54207924.0, "step": 3335 }, { "entropy": 4.919010305404663, "epoch": 2.7312883435582824, "grad_norm": 0.484375, "learning_rate": 0.00045344695800267977, "loss": 4.7746, "mean_token_accuracy": 0.24240046888589858, "num_tokens": 54288113.0, "step": 3340 }, { "entropy": 4.939077234268188, "epoch": 2.7353783231083844, "grad_norm": 0.50390625, "learning_rate": 0.00045325491884322204, "loss": 4.734, "mean_token_accuracy": 0.24732884913682937, "num_tokens": 54369755.0, "step": 3345 }, { "entropy": 4.9576647758483885, "epoch": 2.7394683026584867, "grad_norm": 0.5078125, "learning_rate": 0.0004530625303058145, "loss": 4.8101, "mean_token_accuracy": 0.24170378893613814, "num_tokens": 54451390.0, "step": 3350 }, { "entropy": 4.966122627258301, "epoch": 2.743558282208589, "grad_norm": 0.515625, "learning_rate": 0.00045286979276753673, "loss": 4.7974, "mean_token_accuracy": 0.24010733664035797, "num_tokens": 54532733.0, "step": 3355 }, { "entropy": 4.934459018707275, "epoch": 2.747648261758691, "grad_norm": 0.484375, "learning_rate": 0.0004526767066061526, "loss": 4.779, "mean_token_accuracy": 0.24546415209770203, "num_tokens": 54614316.0, "step": 3360 }, { "entropy": 4.893406677246094, "epoch": 2.7517382413087934, "grad_norm": 0.5078125, "learning_rate": 0.000452483272200109, "loss": 4.7138, "mean_token_accuracy": 0.2523727759718895, "num_tokens": 54695892.0, "step": 3365 }, { "entropy": 4.938384246826172, "epoch": 2.755828220858896, "grad_norm": 0.5078125, "learning_rate": 0.00045228948992853576, "loss": 4.7534, "mean_token_accuracy": 0.2454248383641243, "num_tokens": 54776796.0, "step": 3370 }, { "entropy": 4.958254671096801, "epoch": 2.7599182004089977, "grad_norm": 0.49609375, "learning_rate": 0.0004520953601712441, "loss": 4.8114, "mean_token_accuracy": 0.2433888077735901, "num_tokens": 54858145.0, "step": 3375 }, { "entropy": 4.930399751663208, "epoch": 2.7640081799591, "grad_norm": 0.515625, "learning_rate": 0.00045190088330872654, "loss": 4.7375, "mean_token_accuracy": 0.2487674281001091, "num_tokens": 54939824.0, "step": 3380 }, { "entropy": 4.91034460067749, "epoch": 2.7680981595092025, "grad_norm": 0.51171875, "learning_rate": 0.000451706059722156, "loss": 4.7842, "mean_token_accuracy": 0.2441750794649124, "num_tokens": 55021420.0, "step": 3385 }, { "entropy": 4.943553924560547, "epoch": 2.772188139059305, "grad_norm": 0.51171875, "learning_rate": 0.00045151088979338466, "loss": 4.7527, "mean_token_accuracy": 0.2476951703429222, "num_tokens": 55103038.0, "step": 3390 }, { "entropy": 4.945084857940674, "epoch": 2.776278118609407, "grad_norm": 0.490234375, "learning_rate": 0.00045131537390494394, "loss": 4.7787, "mean_token_accuracy": 0.24442900270223616, "num_tokens": 55184715.0, "step": 3395 }, { "entropy": 4.926125335693359, "epoch": 2.780368098159509, "grad_norm": 0.5078125, "learning_rate": 0.0004511195124400429, "loss": 4.7473, "mean_token_accuracy": 0.2424149304628372, "num_tokens": 55266252.0, "step": 3400 }, { "entropy": 4.93266191482544, "epoch": 2.7844580777096115, "grad_norm": 0.498046875, "learning_rate": 0.0004509233057825684, "loss": 4.7619, "mean_token_accuracy": 0.247728630900383, "num_tokens": 55347116.0, "step": 3405 }, { "entropy": 4.8720625877380375, "epoch": 2.788548057259714, "grad_norm": 0.5078125, "learning_rate": 0.0004507267543170835, "loss": 4.6925, "mean_token_accuracy": 0.2496568650007248, "num_tokens": 55428740.0, "step": 3410 }, { "entropy": 4.88042140007019, "epoch": 2.792638036809816, "grad_norm": 0.50390625, "learning_rate": 0.0004505298584288272, "loss": 4.7198, "mean_token_accuracy": 0.24870816618204117, "num_tokens": 55510048.0, "step": 3415 }, { "entropy": 4.92451286315918, "epoch": 2.796728016359918, "grad_norm": 0.5234375, "learning_rate": 0.0004503326185037135, "loss": 4.7455, "mean_token_accuracy": 0.2467379868030548, "num_tokens": 55590919.0, "step": 3420 }, { "entropy": 4.920601224899292, "epoch": 2.8008179959100206, "grad_norm": 0.5234375, "learning_rate": 0.000450135034928331, "loss": 4.7515, "mean_token_accuracy": 0.24690559208393098, "num_tokens": 55672182.0, "step": 3425 }, { "entropy": 4.880917549133301, "epoch": 2.8049079754601225, "grad_norm": 0.494140625, "learning_rate": 0.0004499371080899415, "loss": 4.7019, "mean_token_accuracy": 0.24874116778373717, "num_tokens": 55753771.0, "step": 3430 }, { "entropy": 4.877732753753662, "epoch": 2.808997955010225, "grad_norm": 0.515625, "learning_rate": 0.0004497388383764796, "loss": 4.6969, "mean_token_accuracy": 0.24868603199720382, "num_tokens": 55835422.0, "step": 3435 }, { "entropy": 4.900001955032349, "epoch": 2.8130879345603272, "grad_norm": 0.515625, "learning_rate": 0.0004495402261765523, "loss": 4.7474, "mean_token_accuracy": 0.24317479878664017, "num_tokens": 55916463.0, "step": 3440 }, { "entropy": 4.923697471618652, "epoch": 2.817177914110429, "grad_norm": 0.490234375, "learning_rate": 0.0004493412718794375, "loss": 4.7807, "mean_token_accuracy": 0.24435440748929976, "num_tokens": 55997524.0, "step": 3445 }, { "entropy": 4.922792673110962, "epoch": 2.8212678936605315, "grad_norm": 0.482421875, "learning_rate": 0.0004491419758750838, "loss": 4.7425, "mean_token_accuracy": 0.24868796318769454, "num_tokens": 56078972.0, "step": 3450 }, { "entropy": 4.937544965744019, "epoch": 2.825357873210634, "grad_norm": 0.515625, "learning_rate": 0.0004489423385541093, "loss": 4.7919, "mean_token_accuracy": 0.24287900775671006, "num_tokens": 56160208.0, "step": 3455 }, { "entropy": 4.899245119094848, "epoch": 2.8294478527607363, "grad_norm": 0.5234375, "learning_rate": 0.0004487423603078014, "loss": 4.7355, "mean_token_accuracy": 0.24870974570512772, "num_tokens": 56241665.0, "step": 3460 }, { "entropy": 4.9333264350891115, "epoch": 2.8335378323108387, "grad_norm": 0.5078125, "learning_rate": 0.00044854204152811565, "loss": 4.7579, "mean_token_accuracy": 0.2459481880068779, "num_tokens": 56322397.0, "step": 3465 }, { "entropy": 4.907567071914673, "epoch": 2.8376278118609406, "grad_norm": 0.49609375, "learning_rate": 0.00044834138260767487, "loss": 4.7203, "mean_token_accuracy": 0.24710575640201568, "num_tokens": 56404092.0, "step": 3470 }, { "entropy": 4.952750635147095, "epoch": 2.841717791411043, "grad_norm": 0.5390625, "learning_rate": 0.0004481403839397685, "loss": 4.7708, "mean_token_accuracy": 0.24666679054498672, "num_tokens": 56485340.0, "step": 3475 }, { "entropy": 4.908222913742065, "epoch": 2.8458077709611453, "grad_norm": 0.515625, "learning_rate": 0.0004479390459183522, "loss": 4.7412, "mean_token_accuracy": 0.2461302787065506, "num_tokens": 56566871.0, "step": 3480 }, { "entropy": 4.946573972702026, "epoch": 2.8498977505112473, "grad_norm": 0.5078125, "learning_rate": 0.0004477373689380466, "loss": 4.7592, "mean_token_accuracy": 0.24430133253335953, "num_tokens": 56648512.0, "step": 3485 }, { "entropy": 4.860173082351684, "epoch": 2.8539877300613496, "grad_norm": 0.5078125, "learning_rate": 0.00044753535339413654, "loss": 4.7078, "mean_token_accuracy": 0.2500722914934158, "num_tokens": 56729632.0, "step": 3490 }, { "entropy": 4.926275682449341, "epoch": 2.858077709611452, "grad_norm": 0.5, "learning_rate": 0.0004473329996825706, "loss": 4.7181, "mean_token_accuracy": 0.24877085983753205, "num_tokens": 56810795.0, "step": 3495 }, { "entropy": 4.869659948348999, "epoch": 2.862167689161554, "grad_norm": 0.51171875, "learning_rate": 0.0004471303081999602, "loss": 4.7146, "mean_token_accuracy": 0.2502000331878662, "num_tokens": 56892233.0, "step": 3500 }, { "epoch": 2.862167689161554, "eval_entropy": 4.7275595075835595, "eval_loss": 4.815591335296631, "eval_mean_token_accuracy": 0.24654408078281115, "eval_num_tokens": 56892233.0, "eval_runtime": 90.0698, "eval_samples_per_second": 45.931, "eval_steps_per_second": 5.751, "step": 3500 }, { "entropy": 4.9001336097717285, "epoch": 2.8662576687116563, "grad_norm": 0.490234375, "learning_rate": 0.00044692727934357866, "loss": 4.7394, "mean_token_accuracy": 0.24455972760915756, "num_tokens": 56973240.0, "step": 3505 }, { "entropy": 4.9066328525543215, "epoch": 2.8703476482617587, "grad_norm": 0.50390625, "learning_rate": 0.00044672391351136074, "loss": 4.7291, "mean_token_accuracy": 0.2466620996594429, "num_tokens": 57054415.0, "step": 3510 }, { "entropy": 4.882565498352051, "epoch": 2.874437627811861, "grad_norm": 0.498046875, "learning_rate": 0.0004465202111019013, "loss": 4.731, "mean_token_accuracy": 0.24657192677259446, "num_tokens": 57136075.0, "step": 3515 }, { "entropy": 4.909894132614136, "epoch": 2.8785276073619634, "grad_norm": 0.50390625, "learning_rate": 0.0004463161725144554, "loss": 4.7317, "mean_token_accuracy": 0.24649485051631928, "num_tokens": 57217679.0, "step": 3520 }, { "entropy": 4.8771239757537845, "epoch": 2.8826175869120654, "grad_norm": 0.49609375, "learning_rate": 0.00044611179814893655, "loss": 4.7075, "mean_token_accuracy": 0.2509648188948631, "num_tokens": 57299345.0, "step": 3525 }, { "entropy": 4.863950586318969, "epoch": 2.8867075664621678, "grad_norm": 0.50390625, "learning_rate": 0.00044590708840591665, "loss": 4.7163, "mean_token_accuracy": 0.24861733615398407, "num_tokens": 57380908.0, "step": 3530 }, { "entropy": 4.903891706466675, "epoch": 2.89079754601227, "grad_norm": 0.51171875, "learning_rate": 0.00044570204368662476, "loss": 4.6824, "mean_token_accuracy": 0.24977755695581436, "num_tokens": 57462141.0, "step": 3535 }, { "entropy": 4.850998973846435, "epoch": 2.894887525562372, "grad_norm": 0.49609375, "learning_rate": 0.0004454966643929467, "loss": 4.7137, "mean_token_accuracy": 0.25013542324304583, "num_tokens": 57543731.0, "step": 3540 }, { "entropy": 4.903712511062622, "epoch": 2.8989775051124744, "grad_norm": 0.5078125, "learning_rate": 0.00044529095092742387, "loss": 4.7292, "mean_token_accuracy": 0.24768730700016023, "num_tokens": 57624827.0, "step": 3545 }, { "entropy": 4.910838317871094, "epoch": 2.903067484662577, "grad_norm": 0.494140625, "learning_rate": 0.00044508490369325277, "loss": 4.7596, "mean_token_accuracy": 0.2449783906340599, "num_tokens": 57706512.0, "step": 3550 }, { "entropy": 4.942079830169678, "epoch": 2.9071574642126787, "grad_norm": 0.48828125, "learning_rate": 0.000444878523094284, "loss": 4.7307, "mean_token_accuracy": 0.24698111712932586, "num_tokens": 57787862.0, "step": 3555 }, { "entropy": 4.8942005157470705, "epoch": 2.911247443762781, "grad_norm": 0.5, "learning_rate": 0.00044467180953502154, "loss": 4.7676, "mean_token_accuracy": 0.24507140070199968, "num_tokens": 57869362.0, "step": 3560 }, { "entropy": 4.94478669166565, "epoch": 2.9153374233128835, "grad_norm": 0.5, "learning_rate": 0.000444464763420622, "loss": 4.7569, "mean_token_accuracy": 0.2459695115685463, "num_tokens": 57950883.0, "step": 3565 }, { "entropy": 4.853419542312622, "epoch": 2.9194274028629854, "grad_norm": 0.5, "learning_rate": 0.00044425738515689383, "loss": 4.6887, "mean_token_accuracy": 0.24965947568416597, "num_tokens": 58031795.0, "step": 3570 }, { "entropy": 4.8510504245758055, "epoch": 2.923517382413088, "grad_norm": 0.5078125, "learning_rate": 0.0004440496751502965, "loss": 4.6866, "mean_token_accuracy": 0.25439983904361724, "num_tokens": 58112914.0, "step": 3575 }, { "entropy": 4.832573795318604, "epoch": 2.92760736196319, "grad_norm": 0.50390625, "learning_rate": 0.00044384163380793956, "loss": 4.6834, "mean_token_accuracy": 0.25119674652814866, "num_tokens": 58194485.0, "step": 3580 }, { "entropy": 4.902034521102905, "epoch": 2.9316973415132925, "grad_norm": 0.515625, "learning_rate": 0.00044363326153758215, "loss": 4.7272, "mean_token_accuracy": 0.25104987621307373, "num_tokens": 58276155.0, "step": 3585 }, { "entropy": 4.9184857368469235, "epoch": 2.935787321063395, "grad_norm": 0.5078125, "learning_rate": 0.0004434245587476319, "loss": 4.7419, "mean_token_accuracy": 0.2459889143705368, "num_tokens": 58357231.0, "step": 3590 }, { "entropy": 4.890280866622925, "epoch": 2.939877300613497, "grad_norm": 0.48828125, "learning_rate": 0.0004432155258471443, "loss": 4.7043, "mean_token_accuracy": 0.24841273874044417, "num_tokens": 58438483.0, "step": 3595 }, { "entropy": 4.8938031673431395, "epoch": 2.943967280163599, "grad_norm": 0.50390625, "learning_rate": 0.0004430061632458219, "loss": 4.7651, "mean_token_accuracy": 0.24590953588485717, "num_tokens": 58520079.0, "step": 3600 }, { "entropy": 4.940750312805176, "epoch": 2.9480572597137016, "grad_norm": 0.50390625, "learning_rate": 0.00044279647135401335, "loss": 4.7417, "mean_token_accuracy": 0.24489525258541106, "num_tokens": 58601729.0, "step": 3605 }, { "entropy": 4.902005815505982, "epoch": 2.9521472392638035, "grad_norm": 0.5, "learning_rate": 0.0004425864505827128, "loss": 4.7543, "mean_token_accuracy": 0.24597787708044053, "num_tokens": 58682614.0, "step": 3610 }, { "entropy": 4.9104164123535154, "epoch": 2.956237218813906, "grad_norm": 0.51171875, "learning_rate": 0.00044237610134355895, "loss": 4.7383, "mean_token_accuracy": 0.24709518551826476, "num_tokens": 58763369.0, "step": 3615 }, { "entropy": 4.881744480133056, "epoch": 2.9603271983640083, "grad_norm": 0.5078125, "learning_rate": 0.00044216542404883425, "loss": 4.7049, "mean_token_accuracy": 0.24963617622852324, "num_tokens": 58844285.0, "step": 3620 }, { "entropy": 4.902839708328247, "epoch": 2.96441717791411, "grad_norm": 0.50390625, "learning_rate": 0.0004419544191114644, "loss": 4.7337, "mean_token_accuracy": 0.24774347245693207, "num_tokens": 58925876.0, "step": 3625 }, { "entropy": 4.901867341995239, "epoch": 2.9685071574642126, "grad_norm": 0.52734375, "learning_rate": 0.00044174308694501686, "loss": 4.7534, "mean_token_accuracy": 0.24608446210622786, "num_tokens": 59006917.0, "step": 3630 }, { "entropy": 4.907237386703491, "epoch": 2.972597137014315, "grad_norm": 0.50390625, "learning_rate": 0.0004415314279637008, "loss": 4.7479, "mean_token_accuracy": 0.2480717420578003, "num_tokens": 59088320.0, "step": 3635 }, { "entropy": 4.887152671813965, "epoch": 2.9766871165644173, "grad_norm": 0.490234375, "learning_rate": 0.00044131944258236585, "loss": 4.7194, "mean_token_accuracy": 0.25277585089206694, "num_tokens": 59169548.0, "step": 3640 }, { "entropy": 4.924459028244018, "epoch": 2.9807770961145197, "grad_norm": 0.5078125, "learning_rate": 0.00044110713121650114, "loss": 4.7535, "mean_token_accuracy": 0.245216403901577, "num_tokens": 59251161.0, "step": 3645 }, { "entropy": 4.87201189994812, "epoch": 2.9848670756646216, "grad_norm": 0.482421875, "learning_rate": 0.0004408944942822351, "loss": 4.7027, "mean_token_accuracy": 0.24860555529594422, "num_tokens": 59331945.0, "step": 3650 }, { "entropy": 4.893537950515747, "epoch": 2.988957055214724, "grad_norm": 0.494140625, "learning_rate": 0.0004406815321963341, "loss": 4.7009, "mean_token_accuracy": 0.24918642044067382, "num_tokens": 59413140.0, "step": 3655 }, { "entropy": 4.946266746520996, "epoch": 2.9930470347648264, "grad_norm": 0.494140625, "learning_rate": 0.00044046824537620174, "loss": 4.7808, "mean_token_accuracy": 0.24528264999389648, "num_tokens": 59494724.0, "step": 3660 }, { "entropy": 4.889382457733154, "epoch": 2.9971370143149283, "grad_norm": 0.515625, "learning_rate": 0.00044025463423987823, "loss": 4.7288, "mean_token_accuracy": 0.2447082743048668, "num_tokens": 59576451.0, "step": 3665 }, { "entropy": 4.8749978807237415, "epoch": 3.0008179959100203, "grad_norm": 0.53125, "learning_rate": 0.00044004069920603936, "loss": 4.7077, "mean_token_accuracy": 0.24972608354356554, "num_tokens": 59642836.0, "step": 3670 }, { "entropy": 4.882791948318482, "epoch": 3.0049079754601227, "grad_norm": 0.51953125, "learning_rate": 0.0004398264406939957, "loss": 4.5786, "mean_token_accuracy": 0.25828482657670976, "num_tokens": 59724122.0, "step": 3675 }, { "entropy": 4.865776491165161, "epoch": 3.008997955010225, "grad_norm": 0.51171875, "learning_rate": 0.00043961185912369196, "loss": 4.5904, "mean_token_accuracy": 0.2582633912563324, "num_tokens": 59805239.0, "step": 3680 }, { "entropy": 4.896941471099853, "epoch": 3.0130879345603274, "grad_norm": 0.51171875, "learning_rate": 0.00043939695491570596, "loss": 4.6112, "mean_token_accuracy": 0.25508000403642656, "num_tokens": 59886843.0, "step": 3685 }, { "entropy": 4.8918993949890135, "epoch": 3.0171779141104293, "grad_norm": 0.5, "learning_rate": 0.000439181728491248, "loss": 4.6019, "mean_token_accuracy": 0.25431698113679885, "num_tokens": 59968333.0, "step": 3690 }, { "entropy": 4.910123872756958, "epoch": 3.0212678936605317, "grad_norm": 0.51171875, "learning_rate": 0.0004389661802721596, "loss": 4.626, "mean_token_accuracy": 0.25652379393577573, "num_tokens": 60049731.0, "step": 3695 }, { "entropy": 4.839553356170654, "epoch": 3.025357873210634, "grad_norm": 0.50390625, "learning_rate": 0.0004387503106809135, "loss": 4.5666, "mean_token_accuracy": 0.26130115389823916, "num_tokens": 60131292.0, "step": 3700 }, { "entropy": 4.840274953842163, "epoch": 3.029447852760736, "grad_norm": 0.5, "learning_rate": 0.0004385341201406119, "loss": 4.5379, "mean_token_accuracy": 0.2599867403507233, "num_tokens": 60211964.0, "step": 3705 }, { "entropy": 4.84863429069519, "epoch": 3.0335378323108384, "grad_norm": 0.5078125, "learning_rate": 0.00043831760907498627, "loss": 4.602, "mean_token_accuracy": 0.2575831949710846, "num_tokens": 60293380.0, "step": 3710 }, { "entropy": 4.897417449951172, "epoch": 3.0376278118609408, "grad_norm": 0.5, "learning_rate": 0.00043810077790839614, "loss": 4.6015, "mean_token_accuracy": 0.25631198436021807, "num_tokens": 60375035.0, "step": 3715 }, { "entropy": 4.854046630859375, "epoch": 3.041717791411043, "grad_norm": 0.51171875, "learning_rate": 0.00043788362706582865, "loss": 4.5914, "mean_token_accuracy": 0.25678600668907164, "num_tokens": 60456342.0, "step": 3720 }, { "entropy": 4.872200345993042, "epoch": 3.045807770961145, "grad_norm": 0.51171875, "learning_rate": 0.0004376661569728973, "loss": 4.5745, "mean_token_accuracy": 0.2561569198966026, "num_tokens": 60537538.0, "step": 3725 }, { "entropy": 4.86928539276123, "epoch": 3.0498977505112475, "grad_norm": 0.51953125, "learning_rate": 0.0004374483680558415, "loss": 4.614, "mean_token_accuracy": 0.25436386913061143, "num_tokens": 60618733.0, "step": 3730 }, { "entropy": 4.836661958694458, "epoch": 3.05398773006135, "grad_norm": 0.51953125, "learning_rate": 0.0004372302607415254, "loss": 4.5442, "mean_token_accuracy": 0.26196159720420836, "num_tokens": 60699879.0, "step": 3735 }, { "entropy": 4.814663362503052, "epoch": 3.0580777096114518, "grad_norm": 0.515625, "learning_rate": 0.0004370118354574373, "loss": 4.5589, "mean_token_accuracy": 0.2581883788108826, "num_tokens": 60781554.0, "step": 3740 }, { "entropy": 4.904482507705689, "epoch": 3.062167689161554, "grad_norm": 0.51953125, "learning_rate": 0.00043679309263168866, "loss": 4.6208, "mean_token_accuracy": 0.25658206045627596, "num_tokens": 60863125.0, "step": 3745 }, { "entropy": 4.844352102279663, "epoch": 3.0662576687116565, "grad_norm": 0.515625, "learning_rate": 0.0004365740326930134, "loss": 4.5714, "mean_token_accuracy": 0.26040266305208204, "num_tokens": 60944411.0, "step": 3750 }, { "entropy": 4.852404260635376, "epoch": 3.070347648261759, "grad_norm": 0.51953125, "learning_rate": 0.00043635465607076685, "loss": 4.5767, "mean_token_accuracy": 0.2568571656942368, "num_tokens": 61025858.0, "step": 3755 }, { "entropy": 4.835520648956299, "epoch": 3.074437627811861, "grad_norm": 0.50390625, "learning_rate": 0.00043613496319492505, "loss": 4.5692, "mean_token_accuracy": 0.2570141404867172, "num_tokens": 61107548.0, "step": 3760 }, { "entropy": 4.8887364864349365, "epoch": 3.078527607361963, "grad_norm": 0.52734375, "learning_rate": 0.00043591495449608414, "loss": 4.5711, "mean_token_accuracy": 0.2586534917354584, "num_tokens": 61188856.0, "step": 3765 }, { "entropy": 4.848718118667603, "epoch": 3.0826175869120656, "grad_norm": 0.51171875, "learning_rate": 0.00043569463040545893, "loss": 4.6074, "mean_token_accuracy": 0.25696409195661546, "num_tokens": 61270038.0, "step": 3770 }, { "entropy": 4.8726245880126955, "epoch": 3.0867075664621675, "grad_norm": 0.50390625, "learning_rate": 0.00043547399135488265, "loss": 4.5865, "mean_token_accuracy": 0.25920682996511457, "num_tokens": 61351249.0, "step": 3775 }, { "entropy": 4.8384361743927, "epoch": 3.09079754601227, "grad_norm": 0.49609375, "learning_rate": 0.00043525303777680564, "loss": 4.5989, "mean_token_accuracy": 0.25728286504745485, "num_tokens": 61432214.0, "step": 3780 }, { "entropy": 4.820977210998535, "epoch": 3.0948875255623722, "grad_norm": 0.5078125, "learning_rate": 0.0004350317701042949, "loss": 4.5562, "mean_token_accuracy": 0.26111441403627395, "num_tokens": 61513854.0, "step": 3785 }, { "entropy": 4.84068717956543, "epoch": 3.0989775051124746, "grad_norm": 0.5, "learning_rate": 0.0004348101887710329, "loss": 4.5739, "mean_token_accuracy": 0.2578523486852646, "num_tokens": 61595246.0, "step": 3790 }, { "entropy": 4.882541513442993, "epoch": 3.1030674846625765, "grad_norm": 0.52734375, "learning_rate": 0.00043458829421131706, "loss": 4.625, "mean_token_accuracy": 0.25619224905967714, "num_tokens": 61676729.0, "step": 3795 }, { "entropy": 4.895043277740479, "epoch": 3.107157464212679, "grad_norm": 0.51171875, "learning_rate": 0.0004343660868600585, "loss": 4.6202, "mean_token_accuracy": 0.25579540282487867, "num_tokens": 61758176.0, "step": 3800 }, { "entropy": 4.86794114112854, "epoch": 3.1112474437627813, "grad_norm": 0.5234375, "learning_rate": 0.00043414356715278156, "loss": 4.6267, "mean_token_accuracy": 0.25599191933870313, "num_tokens": 61839220.0, "step": 3805 }, { "entropy": 4.872279214859009, "epoch": 3.1153374233128837, "grad_norm": 0.5390625, "learning_rate": 0.0004339207355256228, "loss": 4.5973, "mean_token_accuracy": 0.2587134584784508, "num_tokens": 61920792.0, "step": 3810 }, { "entropy": 4.804610013961792, "epoch": 3.1194274028629856, "grad_norm": 0.5, "learning_rate": 0.00043369759241533, "loss": 4.5674, "mean_token_accuracy": 0.25854517668485644, "num_tokens": 62002455.0, "step": 3815 }, { "entropy": 4.850857162475586, "epoch": 3.123517382413088, "grad_norm": 0.515625, "learning_rate": 0.00043347413825926156, "loss": 4.5893, "mean_token_accuracy": 0.258424212038517, "num_tokens": 62083969.0, "step": 3820 }, { "entropy": 4.848945188522339, "epoch": 3.1276073619631903, "grad_norm": 0.50390625, "learning_rate": 0.0004332503734953856, "loss": 4.5553, "mean_token_accuracy": 0.25818339437246324, "num_tokens": 62165654.0, "step": 3825 }, { "entropy": 4.817874908447266, "epoch": 3.1316973415132923, "grad_norm": 0.5078125, "learning_rate": 0.0004330262985622789, "loss": 4.6087, "mean_token_accuracy": 0.25667659640312196, "num_tokens": 62246872.0, "step": 3830 }, { "entropy": 4.881760168075561, "epoch": 3.1357873210633946, "grad_norm": 0.515625, "learning_rate": 0.0004328019138991261, "loss": 4.6012, "mean_token_accuracy": 0.258481428027153, "num_tokens": 62327983.0, "step": 3835 }, { "entropy": 4.841020107269287, "epoch": 3.139877300613497, "grad_norm": 0.51171875, "learning_rate": 0.0004325772199457191, "loss": 4.5648, "mean_token_accuracy": 0.25901042073965075, "num_tokens": 62409086.0, "step": 3840 }, { "entropy": 4.8249499797821045, "epoch": 3.1439672801635994, "grad_norm": 0.515625, "learning_rate": 0.0004323522171424559, "loss": 4.569, "mean_token_accuracy": 0.25887053906917573, "num_tokens": 62490614.0, "step": 3845 }, { "entropy": 4.868269062042236, "epoch": 3.1480572597137013, "grad_norm": 0.52734375, "learning_rate": 0.00043212690593033993, "loss": 4.6061, "mean_token_accuracy": 0.25853896886110306, "num_tokens": 62572353.0, "step": 3850 }, { "entropy": 4.83723554611206, "epoch": 3.1521472392638037, "grad_norm": 0.51171875, "learning_rate": 0.00043190128675097897, "loss": 4.5647, "mean_token_accuracy": 0.26210694164037707, "num_tokens": 62652553.0, "step": 3855 }, { "entropy": 4.830539703369141, "epoch": 3.156237218813906, "grad_norm": 0.51953125, "learning_rate": 0.00043167536004658436, "loss": 4.5995, "mean_token_accuracy": 0.25579892098903656, "num_tokens": 62734013.0, "step": 3860 }, { "entropy": 4.8751708984375, "epoch": 3.160327198364008, "grad_norm": 0.515625, "learning_rate": 0.00043144912625997045, "loss": 4.5829, "mean_token_accuracy": 0.2591325655579567, "num_tokens": 62815208.0, "step": 3865 }, { "entropy": 4.804668188095093, "epoch": 3.1644171779141104, "grad_norm": 0.5078125, "learning_rate": 0.0004312225858345533, "loss": 4.5666, "mean_token_accuracy": 0.2606279879808426, "num_tokens": 62896478.0, "step": 3870 }, { "entropy": 4.837849426269531, "epoch": 3.1685071574642127, "grad_norm": 0.515625, "learning_rate": 0.00043099573921434993, "loss": 4.5906, "mean_token_accuracy": 0.25853571593761443, "num_tokens": 62977635.0, "step": 3875 }, { "entropy": 4.846862840652466, "epoch": 3.172597137014315, "grad_norm": 0.515625, "learning_rate": 0.00043076858684397765, "loss": 4.6088, "mean_token_accuracy": 0.2569206848740578, "num_tokens": 63059344.0, "step": 3880 }, { "entropy": 4.859775018692017, "epoch": 3.176687116564417, "grad_norm": 0.52734375, "learning_rate": 0.00043054112916865277, "loss": 4.5851, "mean_token_accuracy": 0.2564369171857834, "num_tokens": 63140625.0, "step": 3885 }, { "entropy": 4.843816614151001, "epoch": 3.1807770961145194, "grad_norm": 0.51171875, "learning_rate": 0.0004303133666341904, "loss": 4.5822, "mean_token_accuracy": 0.2590786501765251, "num_tokens": 63221903.0, "step": 3890 }, { "entropy": 4.792927694320679, "epoch": 3.184867075664622, "grad_norm": 0.50390625, "learning_rate": 0.00043008529968700283, "loss": 4.54, "mean_token_accuracy": 0.258416585624218, "num_tokens": 63303066.0, "step": 3895 }, { "entropy": 4.826917934417724, "epoch": 3.1889570552147237, "grad_norm": 0.5234375, "learning_rate": 0.00042985692877409913, "loss": 4.5675, "mean_token_accuracy": 0.2623569816350937, "num_tokens": 63384681.0, "step": 3900 }, { "entropy": 4.832444763183593, "epoch": 3.193047034764826, "grad_norm": 0.5546875, "learning_rate": 0.0004296282543430841, "loss": 4.5998, "mean_token_accuracy": 0.2566315457224846, "num_tokens": 63466240.0, "step": 3905 }, { "entropy": 4.8148548126220705, "epoch": 3.1971370143149285, "grad_norm": 0.53125, "learning_rate": 0.0004293992768421575, "loss": 4.5699, "mean_token_accuracy": 0.25920642018318174, "num_tokens": 63547330.0, "step": 3910 }, { "entropy": 4.824674797058106, "epoch": 3.201226993865031, "grad_norm": 0.53125, "learning_rate": 0.00042916999672011304, "loss": 4.5728, "mean_token_accuracy": 0.26058426797389983, "num_tokens": 63628466.0, "step": 3915 }, { "entropy": 4.819450807571411, "epoch": 3.2053169734151328, "grad_norm": 0.51171875, "learning_rate": 0.00042894041442633764, "loss": 4.5733, "mean_token_accuracy": 0.25970950722694397, "num_tokens": 63709600.0, "step": 3920 }, { "entropy": 4.8056408882141115, "epoch": 3.209406952965235, "grad_norm": 0.50390625, "learning_rate": 0.0004287105304108103, "loss": 4.5415, "mean_token_accuracy": 0.262380450963974, "num_tokens": 63791288.0, "step": 3925 }, { "entropy": 4.870853424072266, "epoch": 3.2134969325153375, "grad_norm": 0.5234375, "learning_rate": 0.00042848034512410157, "loss": 4.6113, "mean_token_accuracy": 0.2560029998421669, "num_tokens": 63872705.0, "step": 3930 }, { "entropy": 4.82818603515625, "epoch": 3.21758691206544, "grad_norm": 0.5078125, "learning_rate": 0.0004282498590173724, "loss": 4.5926, "mean_token_accuracy": 0.2590706303715706, "num_tokens": 63954412.0, "step": 3935 }, { "entropy": 4.856594133377075, "epoch": 3.221676891615542, "grad_norm": 0.51953125, "learning_rate": 0.00042801907254237344, "loss": 4.5734, "mean_token_accuracy": 0.2605728358030319, "num_tokens": 64036068.0, "step": 3940 }, { "entropy": 4.811234331130981, "epoch": 3.225766871165644, "grad_norm": 0.50390625, "learning_rate": 0.00042778798615144393, "loss": 4.5803, "mean_token_accuracy": 0.26036179810762405, "num_tokens": 64117155.0, "step": 3945 }, { "entropy": 4.784147930145264, "epoch": 3.2298568507157466, "grad_norm": 0.515625, "learning_rate": 0.000427556600297511, "loss": 4.5344, "mean_token_accuracy": 0.26157242357730864, "num_tokens": 64198417.0, "step": 3950 }, { "entropy": 4.819243144989014, "epoch": 3.2339468302658485, "grad_norm": 0.53125, "learning_rate": 0.00042732491543408886, "loss": 4.5854, "mean_token_accuracy": 0.25791202634572985, "num_tokens": 64280065.0, "step": 3955 }, { "entropy": 4.84360671043396, "epoch": 3.238036809815951, "grad_norm": 0.498046875, "learning_rate": 0.0004270929320152774, "loss": 4.6071, "mean_token_accuracy": 0.2562189131975174, "num_tokens": 64361404.0, "step": 3960 }, { "entropy": 4.823956298828125, "epoch": 3.2421267893660533, "grad_norm": 0.4921875, "learning_rate": 0.0004268606504957622, "loss": 4.5814, "mean_token_accuracy": 0.25823610424995425, "num_tokens": 64442836.0, "step": 3965 }, { "entropy": 4.806665325164795, "epoch": 3.2462167689161556, "grad_norm": 0.53515625, "learning_rate": 0.0004266280713308126, "loss": 4.5906, "mean_token_accuracy": 0.2571432784199715, "num_tokens": 64523176.0, "step": 3970 }, { "entropy": 4.822175121307373, "epoch": 3.2503067484662576, "grad_norm": 0.515625, "learning_rate": 0.0004263951949762817, "loss": 4.5741, "mean_token_accuracy": 0.2581242725253105, "num_tokens": 64604697.0, "step": 3975 }, { "entropy": 4.83758578300476, "epoch": 3.25439672801636, "grad_norm": 0.5078125, "learning_rate": 0.00042616202188860486, "loss": 4.6176, "mean_token_accuracy": 0.2549217641353607, "num_tokens": 64686269.0, "step": 3980 }, { "entropy": 4.814913034439087, "epoch": 3.2584867075664623, "grad_norm": 0.515625, "learning_rate": 0.0004259285525247992, "loss": 4.5615, "mean_token_accuracy": 0.2582083582878113, "num_tokens": 64767855.0, "step": 3985 }, { "entropy": 4.8553155899047855, "epoch": 3.2625766871165642, "grad_norm": 0.51171875, "learning_rate": 0.00042569478734246234, "loss": 4.615, "mean_token_accuracy": 0.2563172996044159, "num_tokens": 64848674.0, "step": 3990 }, { "entropy": 4.814035081863404, "epoch": 3.2666666666666666, "grad_norm": 0.5078125, "learning_rate": 0.00042546072679977196, "loss": 4.5502, "mean_token_accuracy": 0.2594105303287506, "num_tokens": 64929873.0, "step": 3995 }, { "entropy": 4.795870447158814, "epoch": 3.270756646216769, "grad_norm": 0.494140625, "learning_rate": 0.0004252263713554844, "loss": 4.5524, "mean_token_accuracy": 0.26086052060127257, "num_tokens": 65011532.0, "step": 4000 }, { "epoch": 3.270756646216769, "eval_entropy": 4.6499642294806405, "eval_loss": 4.767281532287598, "eval_mean_token_accuracy": 0.251413153117688, "eval_num_tokens": 65011532.0, "eval_runtime": 67.2748, "eval_samples_per_second": 61.494, "eval_steps_per_second": 7.7, "step": 4000 }, { "entropy": 4.826226902008057, "epoch": 3.2748466257668714, "grad_norm": 0.515625, "learning_rate": 0.00042499172146893423, "loss": 4.5901, "mean_token_accuracy": 0.25587351620197296, "num_tokens": 65093167.0, "step": 4005 }, { "entropy": 4.848085069656372, "epoch": 3.2789366053169733, "grad_norm": 0.50390625, "learning_rate": 0.00042475677760003285, "loss": 4.5876, "mean_token_accuracy": 0.2598945677280426, "num_tokens": 65174689.0, "step": 4010 }, { "entropy": 4.837499046325684, "epoch": 3.2830265848670757, "grad_norm": 0.515625, "learning_rate": 0.000424521540209268, "loss": 4.6127, "mean_token_accuracy": 0.25635616630315783, "num_tokens": 65255812.0, "step": 4015 }, { "entropy": 4.788172245025635, "epoch": 3.287116564417178, "grad_norm": 0.54296875, "learning_rate": 0.000424286009757703, "loss": 4.5469, "mean_token_accuracy": 0.2612452819943428, "num_tokens": 65337246.0, "step": 4020 }, { "entropy": 4.877437496185303, "epoch": 3.29120654396728, "grad_norm": 0.51171875, "learning_rate": 0.000424050186706975, "loss": 4.5995, "mean_token_accuracy": 0.2553956612944603, "num_tokens": 65418883.0, "step": 4025 }, { "entropy": 4.8051464557647705, "epoch": 3.2952965235173823, "grad_norm": 0.5234375, "learning_rate": 0.000423814071519295, "loss": 4.5821, "mean_token_accuracy": 0.25973493307828904, "num_tokens": 65500504.0, "step": 4030 }, { "entropy": 4.821094512939453, "epoch": 3.2993865030674847, "grad_norm": 0.5, "learning_rate": 0.0004235776646574467, "loss": 4.5669, "mean_token_accuracy": 0.2582583338022232, "num_tokens": 65581606.0, "step": 4035 }, { "entropy": 4.79600191116333, "epoch": 3.303476482617587, "grad_norm": 0.5078125, "learning_rate": 0.00042334096658478516, "loss": 4.5779, "mean_token_accuracy": 0.2571506381034851, "num_tokens": 65662554.0, "step": 4040 }, { "entropy": 4.8304838180542, "epoch": 3.307566462167689, "grad_norm": 0.515625, "learning_rate": 0.00042310397776523643, "loss": 4.5871, "mean_token_accuracy": 0.2585014432668686, "num_tokens": 65744199.0, "step": 4045 }, { "entropy": 4.852292919158936, "epoch": 3.3116564417177914, "grad_norm": 0.5078125, "learning_rate": 0.0004228666986632962, "loss": 4.5824, "mean_token_accuracy": 0.2611838772892952, "num_tokens": 65825200.0, "step": 4050 }, { "entropy": 4.809878349304199, "epoch": 3.3157464212678938, "grad_norm": 0.5234375, "learning_rate": 0.0004226291297440294, "loss": 4.5729, "mean_token_accuracy": 0.2605692446231842, "num_tokens": 65906836.0, "step": 4055 }, { "entropy": 4.850570821762085, "epoch": 3.319836400817996, "grad_norm": 0.53125, "learning_rate": 0.0004223912714730689, "loss": 4.5879, "mean_token_accuracy": 0.2613891541957855, "num_tokens": 65988235.0, "step": 4060 }, { "entropy": 4.786457586288452, "epoch": 3.323926380368098, "grad_norm": 0.49609375, "learning_rate": 0.0004221531243166145, "loss": 4.573, "mean_token_accuracy": 0.260275799036026, "num_tokens": 66069470.0, "step": 4065 }, { "entropy": 4.839975309371948, "epoch": 3.3280163599182004, "grad_norm": 0.5078125, "learning_rate": 0.00042191468874143244, "loss": 4.593, "mean_token_accuracy": 0.2595179155468941, "num_tokens": 66150057.0, "step": 4070 }, { "entropy": 4.792950677871704, "epoch": 3.332106339468303, "grad_norm": 0.515625, "learning_rate": 0.0004216759652148542, "loss": 4.5518, "mean_token_accuracy": 0.2630715772509575, "num_tokens": 66231651.0, "step": 4075 }, { "entropy": 4.782386779785156, "epoch": 3.3361963190184047, "grad_norm": 0.51171875, "learning_rate": 0.00042143695420477565, "loss": 4.5393, "mean_token_accuracy": 0.25993366837501525, "num_tokens": 66312168.0, "step": 4080 }, { "entropy": 4.800601196289063, "epoch": 3.340286298568507, "grad_norm": 0.54296875, "learning_rate": 0.0004211976561796562, "loss": 4.569, "mean_token_accuracy": 0.2598921149969101, "num_tokens": 66393328.0, "step": 4085 }, { "entropy": 4.793715906143189, "epoch": 3.3443762781186095, "grad_norm": 0.49609375, "learning_rate": 0.0004209580716085174, "loss": 4.5153, "mean_token_accuracy": 0.2616371840238571, "num_tokens": 66474543.0, "step": 4090 }, { "entropy": 4.785206413269043, "epoch": 3.3484662576687114, "grad_norm": 0.50390625, "learning_rate": 0.000420718200960943, "loss": 4.5484, "mean_token_accuracy": 0.2620126724243164, "num_tokens": 66556164.0, "step": 4095 }, { "entropy": 4.825778818130493, "epoch": 3.352556237218814, "grad_norm": 0.515625, "learning_rate": 0.0004204780447070771, "loss": 4.6168, "mean_token_accuracy": 0.2551762148737907, "num_tokens": 66637856.0, "step": 4100 }, { "entropy": 4.861998844146728, "epoch": 3.356646216768916, "grad_norm": 0.52734375, "learning_rate": 0.0004202376033176236, "loss": 4.633, "mean_token_accuracy": 0.2529298201203346, "num_tokens": 66719463.0, "step": 4105 }, { "entropy": 4.825679206848145, "epoch": 3.3607361963190185, "grad_norm": 0.50390625, "learning_rate": 0.00041999687726384556, "loss": 4.5892, "mean_token_accuracy": 0.2592676505446434, "num_tokens": 66800626.0, "step": 4110 }, { "entropy": 4.856466245651245, "epoch": 3.3648261758691205, "grad_norm": 0.51171875, "learning_rate": 0.0004197558670175635, "loss": 4.6092, "mean_token_accuracy": 0.2560419663786888, "num_tokens": 66881728.0, "step": 4115 }, { "entropy": 4.802723455429077, "epoch": 3.368916155419223, "grad_norm": 0.515625, "learning_rate": 0.0004195145730511554, "loss": 4.5807, "mean_token_accuracy": 0.25446502566337587, "num_tokens": 66962949.0, "step": 4120 }, { "entropy": 4.849379634857177, "epoch": 3.373006134969325, "grad_norm": 0.5078125, "learning_rate": 0.00041927299583755515, "loss": 4.599, "mean_token_accuracy": 0.25521106868982313, "num_tokens": 67043664.0, "step": 4125 }, { "entropy": 4.755837440490723, "epoch": 3.3770961145194276, "grad_norm": 0.5078125, "learning_rate": 0.0004190311358502517, "loss": 4.4954, "mean_token_accuracy": 0.26704823076725004, "num_tokens": 67125240.0, "step": 4130 }, { "entropy": 4.781294298171997, "epoch": 3.3811860940695295, "grad_norm": 0.52734375, "learning_rate": 0.00041878899356328835, "loss": 4.5382, "mean_token_accuracy": 0.26138612926006316, "num_tokens": 67206656.0, "step": 4135 }, { "entropy": 4.819573307037354, "epoch": 3.385276073619632, "grad_norm": 0.515625, "learning_rate": 0.0004185465694512618, "loss": 4.5751, "mean_token_accuracy": 0.25750154703855516, "num_tokens": 67287955.0, "step": 4140 }, { "entropy": 4.758851385116577, "epoch": 3.3893660531697343, "grad_norm": 0.5078125, "learning_rate": 0.00041830386398932087, "loss": 4.5224, "mean_token_accuracy": 0.26153791397809983, "num_tokens": 67369551.0, "step": 4145 }, { "entropy": 4.827676773071289, "epoch": 3.393456032719836, "grad_norm": 0.50390625, "learning_rate": 0.00041806087765316595, "loss": 4.5607, "mean_token_accuracy": 0.2630741000175476, "num_tokens": 67450859.0, "step": 4150 }, { "entropy": 4.716650819778442, "epoch": 3.3975460122699386, "grad_norm": 0.5, "learning_rate": 0.00041781761091904806, "loss": 4.5071, "mean_token_accuracy": 0.26259891390800477, "num_tokens": 67531999.0, "step": 4155 }, { "entropy": 4.868941688537598, "epoch": 3.401635991820041, "grad_norm": 0.51171875, "learning_rate": 0.0004175740642637677, "loss": 4.6115, "mean_token_accuracy": 0.25713407844305036, "num_tokens": 67613136.0, "step": 4160 }, { "entropy": 4.777759599685669, "epoch": 3.4057259713701433, "grad_norm": 0.51171875, "learning_rate": 0.00041733023816467385, "loss": 4.5711, "mean_token_accuracy": 0.2594104573130608, "num_tokens": 67694760.0, "step": 4165 }, { "entropy": 4.844801044464111, "epoch": 3.4098159509202453, "grad_norm": 0.5078125, "learning_rate": 0.0004170861330996635, "loss": 4.5788, "mean_token_accuracy": 0.25622133165597916, "num_tokens": 67775828.0, "step": 4170 }, { "entropy": 4.762381362915039, "epoch": 3.4139059304703476, "grad_norm": 0.515625, "learning_rate": 0.00041684174954718017, "loss": 4.5509, "mean_token_accuracy": 0.2602579802274704, "num_tokens": 67857267.0, "step": 4175 }, { "entropy": 4.77390604019165, "epoch": 3.41799591002045, "grad_norm": 0.49609375, "learning_rate": 0.0004165970879862134, "loss": 4.546, "mean_token_accuracy": 0.26053089201450347, "num_tokens": 67938473.0, "step": 4180 }, { "entropy": 4.795159387588501, "epoch": 3.4220858895705524, "grad_norm": 0.49609375, "learning_rate": 0.0004163521488962975, "loss": 4.5541, "mean_token_accuracy": 0.25897396504879, "num_tokens": 68019607.0, "step": 4185 }, { "entropy": 4.81208553314209, "epoch": 3.4261758691206543, "grad_norm": 0.49609375, "learning_rate": 0.00041610693275751073, "loss": 4.6007, "mean_token_accuracy": 0.2588938668370247, "num_tokens": 68100624.0, "step": 4190 }, { "entropy": 4.814710855484009, "epoch": 3.4302658486707567, "grad_norm": 0.5234375, "learning_rate": 0.0004158614400504743, "loss": 4.5612, "mean_token_accuracy": 0.26241022497415545, "num_tokens": 68181890.0, "step": 4195 }, { "entropy": 4.83739914894104, "epoch": 3.434355828220859, "grad_norm": 0.51953125, "learning_rate": 0.00041561567125635186, "loss": 4.5937, "mean_token_accuracy": 0.2592011481523514, "num_tokens": 68263285.0, "step": 4200 }, { "entropy": 4.797341918945312, "epoch": 3.438445807770961, "grad_norm": 0.5, "learning_rate": 0.00041536962685684753, "loss": 4.5718, "mean_token_accuracy": 0.2615545764565468, "num_tokens": 68344829.0, "step": 4205 }, { "entropy": 4.819930839538574, "epoch": 3.4425357873210634, "grad_norm": 0.5078125, "learning_rate": 0.0004151233073342064, "loss": 4.5758, "mean_token_accuracy": 0.25973864197731017, "num_tokens": 68426409.0, "step": 4210 }, { "entropy": 4.797745323181152, "epoch": 3.4466257668711657, "grad_norm": 0.5, "learning_rate": 0.0004148767131712122, "loss": 4.5797, "mean_token_accuracy": 0.2598946511745453, "num_tokens": 68507673.0, "step": 4215 }, { "entropy": 4.851477193832397, "epoch": 3.4507157464212677, "grad_norm": 0.5234375, "learning_rate": 0.000414629844851187, "loss": 4.637, "mean_token_accuracy": 0.2548974961042404, "num_tokens": 68589063.0, "step": 4220 }, { "entropy": 4.848278522491455, "epoch": 3.45480572597137, "grad_norm": 0.52734375, "learning_rate": 0.0004143827028579907, "loss": 4.6194, "mean_token_accuracy": 0.2558763921260834, "num_tokens": 68670647.0, "step": 4225 }, { "entropy": 4.820701694488525, "epoch": 3.4588957055214724, "grad_norm": 0.5390625, "learning_rate": 0.00041413528767601907, "loss": 4.5777, "mean_token_accuracy": 0.2588019385933876, "num_tokens": 68752386.0, "step": 4230 }, { "entropy": 4.8101091384887695, "epoch": 3.462985685071575, "grad_norm": 0.52734375, "learning_rate": 0.0004138875997902034, "loss": 4.5749, "mean_token_accuracy": 0.26216364204883574, "num_tokens": 68833811.0, "step": 4235 }, { "entropy": 4.790560102462768, "epoch": 3.4670756646216767, "grad_norm": 0.51953125, "learning_rate": 0.0004136396396860099, "loss": 4.5429, "mean_token_accuracy": 0.26364230513572695, "num_tokens": 68915461.0, "step": 4240 }, { "entropy": 4.796834230422974, "epoch": 3.471165644171779, "grad_norm": 0.53125, "learning_rate": 0.00041339140784943775, "loss": 4.5638, "mean_token_accuracy": 0.25868489444255827, "num_tokens": 68997074.0, "step": 4245 }, { "entropy": 4.774380874633789, "epoch": 3.4752556237218815, "grad_norm": 0.4921875, "learning_rate": 0.00041314290476701914, "loss": 4.5271, "mean_token_accuracy": 0.2621817052364349, "num_tokens": 69078398.0, "step": 4250 }, { "entropy": 4.750655221939087, "epoch": 3.479345603271984, "grad_norm": 0.4921875, "learning_rate": 0.0004128941309258177, "loss": 4.5383, "mean_token_accuracy": 0.26234322786331177, "num_tokens": 69159551.0, "step": 4255 }, { "entropy": 4.830879878997803, "epoch": 3.4834355828220858, "grad_norm": 0.50390625, "learning_rate": 0.0004126450868134278, "loss": 4.581, "mean_token_accuracy": 0.25919158607721327, "num_tokens": 69240899.0, "step": 4260 }, { "entropy": 4.794535541534424, "epoch": 3.487525562372188, "grad_norm": 0.5078125, "learning_rate": 0.00041239577291797357, "loss": 4.5818, "mean_token_accuracy": 0.2599661618471146, "num_tokens": 69322358.0, "step": 4265 }, { "entropy": 4.797354316711425, "epoch": 3.4916155419222905, "grad_norm": 0.51171875, "learning_rate": 0.00041214618972810785, "loss": 4.5532, "mean_token_accuracy": 0.2608014464378357, "num_tokens": 69403487.0, "step": 4270 }, { "entropy": 4.786728572845459, "epoch": 3.4957055214723924, "grad_norm": 0.498046875, "learning_rate": 0.00041189633773301145, "loss": 4.5539, "mean_token_accuracy": 0.2595497012138367, "num_tokens": 69485174.0, "step": 4275 }, { "entropy": 4.829294013977051, "epoch": 3.499795501022495, "grad_norm": 0.53125, "learning_rate": 0.00041164621742239166, "loss": 4.5913, "mean_token_accuracy": 0.2583783775568008, "num_tokens": 69566740.0, "step": 4280 }, { "entropy": 4.7829139709472654, "epoch": 3.503885480572597, "grad_norm": 0.49609375, "learning_rate": 0.0004113958292864821, "loss": 4.5444, "mean_token_accuracy": 0.26283992975950243, "num_tokens": 69647415.0, "step": 4285 }, { "entropy": 4.781641721725464, "epoch": 3.507975460122699, "grad_norm": 0.5, "learning_rate": 0.0004111451738160411, "loss": 4.5385, "mean_token_accuracy": 0.26307397782802583, "num_tokens": 69728967.0, "step": 4290 }, { "entropy": 4.808180952072144, "epoch": 3.5120654396728015, "grad_norm": 0.52734375, "learning_rate": 0.00041089425150235093, "loss": 4.5679, "mean_token_accuracy": 0.2615308746695518, "num_tokens": 69809375.0, "step": 4295 }, { "entropy": 4.840615129470825, "epoch": 3.516155419222904, "grad_norm": 0.50390625, "learning_rate": 0.00041064306283721707, "loss": 4.6123, "mean_token_accuracy": 0.2575763791799545, "num_tokens": 69890577.0, "step": 4300 }, { "entropy": 4.795137214660644, "epoch": 3.5202453987730062, "grad_norm": 0.51953125, "learning_rate": 0.0004103916083129669, "loss": 4.5547, "mean_token_accuracy": 0.26073734015226363, "num_tokens": 69972202.0, "step": 4305 }, { "entropy": 4.791865634918213, "epoch": 3.5243353783231086, "grad_norm": 0.494140625, "learning_rate": 0.0004101398884224488, "loss": 4.5652, "mean_token_accuracy": 0.2576068490743637, "num_tokens": 70052439.0, "step": 4310 }, { "entropy": 4.808264112472534, "epoch": 3.5284253578732105, "grad_norm": 0.4921875, "learning_rate": 0.0004098879036590315, "loss": 4.5856, "mean_token_accuracy": 0.2581591635942459, "num_tokens": 70133741.0, "step": 4315 }, { "entropy": 4.789082860946655, "epoch": 3.532515337423313, "grad_norm": 0.4921875, "learning_rate": 0.0004096356545166026, "loss": 4.556, "mean_token_accuracy": 0.2631792426109314, "num_tokens": 70215064.0, "step": 4320 }, { "entropy": 4.780314111709595, "epoch": 3.5366053169734153, "grad_norm": 0.546875, "learning_rate": 0.00040938314148956814, "loss": 4.5402, "mean_token_accuracy": 0.26182631105184556, "num_tokens": 70295665.0, "step": 4325 }, { "entropy": 4.815358638763428, "epoch": 3.540695296523517, "grad_norm": 0.52734375, "learning_rate": 0.00040913036507285124, "loss": 4.5767, "mean_token_accuracy": 0.25662884712219236, "num_tokens": 70377393.0, "step": 4330 }, { "entropy": 4.782622385025024, "epoch": 3.5447852760736196, "grad_norm": 0.52734375, "learning_rate": 0.00040887732576189124, "loss": 4.5423, "mean_token_accuracy": 0.2598027974367142, "num_tokens": 70458753.0, "step": 4335 }, { "entropy": 4.793778419494629, "epoch": 3.548875255623722, "grad_norm": 0.5234375, "learning_rate": 0.00040862402405264275, "loss": 4.553, "mean_token_accuracy": 0.2615514263510704, "num_tokens": 70540074.0, "step": 4340 }, { "entropy": 4.811662769317627, "epoch": 3.552965235173824, "grad_norm": 0.5078125, "learning_rate": 0.0004083704604415748, "loss": 4.5833, "mean_token_accuracy": 0.259107905626297, "num_tokens": 70621803.0, "step": 4345 }, { "entropy": 4.7683923721313475, "epoch": 3.5570552147239263, "grad_norm": 0.5, "learning_rate": 0.00040811663542566974, "loss": 4.5028, "mean_token_accuracy": 0.2634119063615799, "num_tokens": 70703418.0, "step": 4350 }, { "entropy": 4.719976997375488, "epoch": 3.5611451942740286, "grad_norm": 0.498046875, "learning_rate": 0.000407862549502422, "loss": 4.5178, "mean_token_accuracy": 0.25981912910938265, "num_tokens": 70784979.0, "step": 4355 }, { "entropy": 4.814431047439575, "epoch": 3.565235173824131, "grad_norm": 0.49609375, "learning_rate": 0.00040760820316983773, "loss": 4.5896, "mean_token_accuracy": 0.2576896622776985, "num_tokens": 70866664.0, "step": 4360 }, { "entropy": 4.769455766677856, "epoch": 3.5693251533742334, "grad_norm": 0.50390625, "learning_rate": 0.00040735359692643325, "loss": 4.5405, "mean_token_accuracy": 0.260719022154808, "num_tokens": 70947443.0, "step": 4365 }, { "entropy": 4.831691789627075, "epoch": 3.5734151329243353, "grad_norm": 0.51171875, "learning_rate": 0.0004070987312712344, "loss": 4.5984, "mean_token_accuracy": 0.2577139064669609, "num_tokens": 71028620.0, "step": 4370 }, { "entropy": 4.790133380889893, "epoch": 3.5775051124744377, "grad_norm": 0.498046875, "learning_rate": 0.00040684360670377543, "loss": 4.5466, "mean_token_accuracy": 0.26415634751319883, "num_tokens": 71110128.0, "step": 4375 }, { "entropy": 4.788941240310669, "epoch": 3.58159509202454, "grad_norm": 0.490234375, "learning_rate": 0.0004065882237240981, "loss": 4.5605, "mean_token_accuracy": 0.26163511872291567, "num_tokens": 71191763.0, "step": 4380 }, { "entropy": 4.790526914596557, "epoch": 3.585685071574642, "grad_norm": 0.50390625, "learning_rate": 0.0004063325828327506, "loss": 4.5716, "mean_token_accuracy": 0.2600647330284119, "num_tokens": 71273216.0, "step": 4385 }, { "entropy": 4.782167673110962, "epoch": 3.5897750511247444, "grad_norm": 0.51171875, "learning_rate": 0.00040607668453078666, "loss": 4.5753, "mean_token_accuracy": 0.25906757563352584, "num_tokens": 71354319.0, "step": 4390 }, { "entropy": 4.796464920043945, "epoch": 3.5938650306748468, "grad_norm": 0.48828125, "learning_rate": 0.00040582052931976455, "loss": 4.5593, "mean_token_accuracy": 0.2606018155813217, "num_tokens": 71435816.0, "step": 4395 }, { "entropy": 4.797094249725342, "epoch": 3.5979550102249487, "grad_norm": 0.51953125, "learning_rate": 0.00040556411770174605, "loss": 4.5734, "mean_token_accuracy": 0.2601865977048874, "num_tokens": 71517586.0, "step": 4400 }, { "entropy": 4.81096887588501, "epoch": 3.602044989775051, "grad_norm": 0.5078125, "learning_rate": 0.00040530745017929547, "loss": 4.574, "mean_token_accuracy": 0.2590463265776634, "num_tokens": 71598644.0, "step": 4405 }, { "entropy": 4.777805137634277, "epoch": 3.6061349693251534, "grad_norm": 0.50390625, "learning_rate": 0.00040505052725547876, "loss": 4.5556, "mean_token_accuracy": 0.260992094874382, "num_tokens": 71679472.0, "step": 4410 }, { "entropy": 4.814808893203735, "epoch": 3.6102249488752554, "grad_norm": 0.51171875, "learning_rate": 0.0004047933494338623, "loss": 4.578, "mean_token_accuracy": 0.26028763949871064, "num_tokens": 71760144.0, "step": 4415 }, { "entropy": 4.78363037109375, "epoch": 3.6143149284253577, "grad_norm": 0.51953125, "learning_rate": 0.0004045359172185124, "loss": 4.5575, "mean_token_accuracy": 0.264070026576519, "num_tokens": 71841777.0, "step": 4420 }, { "entropy": 4.809151649475098, "epoch": 3.61840490797546, "grad_norm": 0.4921875, "learning_rate": 0.00040427823111399355, "loss": 4.6075, "mean_token_accuracy": 0.25693629682064056, "num_tokens": 71923453.0, "step": 4425 }, { "entropy": 4.855596494674683, "epoch": 3.6224948875255625, "grad_norm": 0.51953125, "learning_rate": 0.00040402029162536817, "loss": 4.6167, "mean_token_accuracy": 0.25590444803237916, "num_tokens": 72005088.0, "step": 4430 }, { "entropy": 4.787191343307495, "epoch": 3.626584867075665, "grad_norm": 0.4921875, "learning_rate": 0.0004037620992581952, "loss": 4.5392, "mean_token_accuracy": 0.2611472249031067, "num_tokens": 72086272.0, "step": 4435 }, { "entropy": 4.775006341934204, "epoch": 3.630674846625767, "grad_norm": 0.515625, "learning_rate": 0.0004035036545185292, "loss": 4.5596, "mean_token_accuracy": 0.2584162220358849, "num_tokens": 72167063.0, "step": 4440 }, { "entropy": 4.813024282455444, "epoch": 3.634764826175869, "grad_norm": 0.5390625, "learning_rate": 0.0004032449579129194, "loss": 4.5756, "mean_token_accuracy": 0.25876259207725527, "num_tokens": 72248774.0, "step": 4445 }, { "entropy": 4.801388502120972, "epoch": 3.6388548057259715, "grad_norm": 0.5, "learning_rate": 0.0004029860099484087, "loss": 4.6021, "mean_token_accuracy": 0.25912988632917405, "num_tokens": 72330348.0, "step": 4450 }, { "entropy": 4.78261981010437, "epoch": 3.6429447852760735, "grad_norm": 0.5078125, "learning_rate": 0.0004027268111325328, "loss": 4.5441, "mean_token_accuracy": 0.26391927897930145, "num_tokens": 72411894.0, "step": 4455 }, { "entropy": 4.801660633087158, "epoch": 3.647034764826176, "grad_norm": 0.51953125, "learning_rate": 0.00040246736197331866, "loss": 4.5844, "mean_token_accuracy": 0.2583887666463852, "num_tokens": 72492861.0, "step": 4460 }, { "entropy": 4.807954406738281, "epoch": 3.651124744376278, "grad_norm": 0.515625, "learning_rate": 0.0004022076629792844, "loss": 4.5464, "mean_token_accuracy": 0.2590378701686859, "num_tokens": 72574576.0, "step": 4465 }, { "entropy": 4.765544319152832, "epoch": 3.65521472392638, "grad_norm": 0.5, "learning_rate": 0.00040194771465943747, "loss": 4.5567, "mean_token_accuracy": 0.26248098015785215, "num_tokens": 72656253.0, "step": 4470 }, { "entropy": 4.764256238937378, "epoch": 3.6593047034764825, "grad_norm": 0.52734375, "learning_rate": 0.0004016875175232741, "loss": 4.5311, "mean_token_accuracy": 0.2637204021215439, "num_tokens": 72737823.0, "step": 4475 }, { "entropy": 4.773712968826294, "epoch": 3.663394683026585, "grad_norm": 0.51171875, "learning_rate": 0.0004014270720807783, "loss": 4.5469, "mean_token_accuracy": 0.26444012820720675, "num_tokens": 72819042.0, "step": 4480 }, { "entropy": 4.81693172454834, "epoch": 3.6674846625766873, "grad_norm": 0.5078125, "learning_rate": 0.0004011663788424205, "loss": 4.5971, "mean_token_accuracy": 0.2611111909151077, "num_tokens": 72900322.0, "step": 4485 }, { "entropy": 4.768061733245849, "epoch": 3.6715746421267896, "grad_norm": 0.515625, "learning_rate": 0.00040090543831915713, "loss": 4.5328, "mean_token_accuracy": 0.26305892169475553, "num_tokens": 72981961.0, "step": 4490 }, { "entropy": 4.8363431930542, "epoch": 3.6756646216768916, "grad_norm": 0.498046875, "learning_rate": 0.00040064425102242916, "loss": 4.6047, "mean_token_accuracy": 0.258725580573082, "num_tokens": 73062791.0, "step": 4495 }, { "entropy": 4.716058349609375, "epoch": 3.679754601226994, "grad_norm": 0.494140625, "learning_rate": 0.00040038281746416105, "loss": 4.4795, "mean_token_accuracy": 0.26656716465950014, "num_tokens": 73144470.0, "step": 4500 }, { "epoch": 3.679754601226994, "eval_entropy": 4.606738321569435, "eval_loss": 4.723547458648682, "eval_mean_token_accuracy": 0.25534805885613193, "eval_num_tokens": 73144470.0, "eval_runtime": 66.0555, "eval_samples_per_second": 62.629, "eval_steps_per_second": 7.842, "step": 4500 }, { "entropy": 4.818124723434448, "epoch": 3.6838445807770963, "grad_norm": 0.498046875, "learning_rate": 0.0004001211381567602, "loss": 4.5945, "mean_token_accuracy": 0.26131935715675353, "num_tokens": 73226083.0, "step": 4505 }, { "entropy": 4.799996709823608, "epoch": 3.6879345603271982, "grad_norm": 0.51171875, "learning_rate": 0.00039985921361311555, "loss": 4.5906, "mean_token_accuracy": 0.2586137741804123, "num_tokens": 73306957.0, "step": 4510 }, { "entropy": 4.7428626537323, "epoch": 3.6920245398773006, "grad_norm": 0.52734375, "learning_rate": 0.00039959704434659683, "loss": 4.5106, "mean_token_accuracy": 0.25890960693359377, "num_tokens": 73387989.0, "step": 4515 }, { "entropy": 4.776598691940308, "epoch": 3.696114519427403, "grad_norm": 0.5078125, "learning_rate": 0.0003993346308710531, "loss": 4.5471, "mean_token_accuracy": 0.26216039806604385, "num_tokens": 73468872.0, "step": 4520 }, { "entropy": 4.783983325958252, "epoch": 3.700204498977505, "grad_norm": 0.51953125, "learning_rate": 0.00039907197370081244, "loss": 4.5573, "mean_token_accuracy": 0.2606941744685173, "num_tokens": 73550194.0, "step": 4525 }, { "entropy": 4.77109694480896, "epoch": 3.7042944785276073, "grad_norm": 0.5078125, "learning_rate": 0.00039880907335068037, "loss": 4.5689, "mean_token_accuracy": 0.26138505935668943, "num_tokens": 73630908.0, "step": 4530 }, { "entropy": 4.778004932403564, "epoch": 3.7083844580777097, "grad_norm": 0.4921875, "learning_rate": 0.000398545930335939, "loss": 4.5364, "mean_token_accuracy": 0.2617111593484879, "num_tokens": 73711741.0, "step": 4535 }, { "entropy": 4.797347831726074, "epoch": 3.7124744376278116, "grad_norm": 0.5, "learning_rate": 0.00039828254517234636, "loss": 4.5951, "mean_token_accuracy": 0.25855430364608767, "num_tokens": 73793207.0, "step": 4540 }, { "entropy": 4.841688776016236, "epoch": 3.716564417177914, "grad_norm": 0.5078125, "learning_rate": 0.0003980189183761346, "loss": 4.6219, "mean_token_accuracy": 0.25859065651893615, "num_tokens": 73874433.0, "step": 4545 }, { "entropy": 4.812349510192871, "epoch": 3.7206543967280163, "grad_norm": 0.490234375, "learning_rate": 0.00039775505046401, "loss": 4.5673, "mean_token_accuracy": 0.259090057015419, "num_tokens": 73955880.0, "step": 4550 }, { "entropy": 4.776797580718994, "epoch": 3.7247443762781187, "grad_norm": 0.5, "learning_rate": 0.000397490941953151, "loss": 4.578, "mean_token_accuracy": 0.26004201620817186, "num_tokens": 74037267.0, "step": 4555 }, { "entropy": 4.73727560043335, "epoch": 3.728834355828221, "grad_norm": 0.4921875, "learning_rate": 0.00039722659336120775, "loss": 4.5155, "mean_token_accuracy": 0.26548105478286743, "num_tokens": 74118386.0, "step": 4560 }, { "entropy": 4.773490333557129, "epoch": 3.732924335378323, "grad_norm": 0.5078125, "learning_rate": 0.00039696200520630115, "loss": 4.5378, "mean_token_accuracy": 0.26412208676338195, "num_tokens": 74199945.0, "step": 4565 }, { "entropy": 4.75936369895935, "epoch": 3.7370143149284254, "grad_norm": 0.5078125, "learning_rate": 0.0003966971780070213, "loss": 4.5326, "mean_token_accuracy": 0.26315838396549224, "num_tokens": 74281295.0, "step": 4570 }, { "entropy": 4.750297498703003, "epoch": 3.7411042944785278, "grad_norm": 0.515625, "learning_rate": 0.0003964321122824271, "loss": 4.5316, "mean_token_accuracy": 0.26392820179462434, "num_tokens": 74362910.0, "step": 4575 }, { "entropy": 4.866172790527344, "epoch": 3.7451942740286297, "grad_norm": 0.52734375, "learning_rate": 0.00039616680855204475, "loss": 4.6473, "mean_token_accuracy": 0.25664288848638533, "num_tokens": 74443894.0, "step": 4580 }, { "entropy": 4.75925464630127, "epoch": 3.749284253578732, "grad_norm": 0.515625, "learning_rate": 0.00039590126733586714, "loss": 4.532, "mean_token_accuracy": 0.26784061789512636, "num_tokens": 74524783.0, "step": 4585 }, { "entropy": 4.741590070724487, "epoch": 3.7533742331288344, "grad_norm": 0.515625, "learning_rate": 0.0003956354891543525, "loss": 4.5093, "mean_token_accuracy": 0.2658362627029419, "num_tokens": 74605703.0, "step": 4590 }, { "entropy": 4.783300971984863, "epoch": 3.7574642126789364, "grad_norm": 0.498046875, "learning_rate": 0.0003953694745284237, "loss": 4.5533, "mean_token_accuracy": 0.2618992134928703, "num_tokens": 74687277.0, "step": 4595 }, { "entropy": 4.8342877388000485, "epoch": 3.7615541922290388, "grad_norm": 0.53125, "learning_rate": 0.00039510322397946665, "loss": 4.6056, "mean_token_accuracy": 0.25989201813936236, "num_tokens": 74768543.0, "step": 4600 }, { "entropy": 4.791373062133789, "epoch": 3.765644171779141, "grad_norm": 0.494140625, "learning_rate": 0.00039483673802932995, "loss": 4.5906, "mean_token_accuracy": 0.2601287066936493, "num_tokens": 74849977.0, "step": 4605 }, { "entropy": 4.816494035720825, "epoch": 3.7697341513292435, "grad_norm": 0.51953125, "learning_rate": 0.00039457001720032364, "loss": 4.5724, "mean_token_accuracy": 0.25816272348165514, "num_tokens": 74931213.0, "step": 4610 }, { "entropy": 4.786914873123169, "epoch": 3.773824130879346, "grad_norm": 0.478515625, "learning_rate": 0.000394303062015218, "loss": 4.5513, "mean_token_accuracy": 0.2606336519122124, "num_tokens": 75012707.0, "step": 4615 }, { "entropy": 4.7502875328063965, "epoch": 3.777914110429448, "grad_norm": 0.5390625, "learning_rate": 0.00039403587299724247, "loss": 4.5716, "mean_token_accuracy": 0.26102918684482573, "num_tokens": 75094027.0, "step": 4620 }, { "entropy": 4.796187829971314, "epoch": 3.78200408997955, "grad_norm": 0.494140625, "learning_rate": 0.00039376845067008514, "loss": 4.5399, "mean_token_accuracy": 0.265798544883728, "num_tokens": 75175436.0, "step": 4625 }, { "entropy": 4.763972520828247, "epoch": 3.7860940695296526, "grad_norm": 0.4921875, "learning_rate": 0.0003935007955578913, "loss": 4.5437, "mean_token_accuracy": 0.2640814185142517, "num_tokens": 75256636.0, "step": 4630 }, { "entropy": 4.73912787437439, "epoch": 3.7901840490797545, "grad_norm": 0.49609375, "learning_rate": 0.0003932329081852623, "loss": 4.5002, "mean_token_accuracy": 0.26314716786146164, "num_tokens": 75337522.0, "step": 4635 }, { "entropy": 4.756170320510864, "epoch": 3.794274028629857, "grad_norm": 0.51171875, "learning_rate": 0.00039296478907725497, "loss": 4.5247, "mean_token_accuracy": 0.262819504737854, "num_tokens": 75418679.0, "step": 4640 }, { "entropy": 4.730951833724975, "epoch": 3.7983640081799592, "grad_norm": 0.474609375, "learning_rate": 0.0003926964387593801, "loss": 4.5197, "mean_token_accuracy": 0.2618684649467468, "num_tokens": 75499930.0, "step": 4645 }, { "entropy": 4.801263618469238, "epoch": 3.802453987730061, "grad_norm": 0.482421875, "learning_rate": 0.00039242785775760186, "loss": 4.5654, "mean_token_accuracy": 0.26215120404958725, "num_tokens": 75581465.0, "step": 4650 }, { "entropy": 4.741362237930298, "epoch": 3.8065439672801635, "grad_norm": 0.51953125, "learning_rate": 0.00039215904659833646, "loss": 4.4925, "mean_token_accuracy": 0.26381993293762207, "num_tokens": 75662746.0, "step": 4655 }, { "entropy": 4.781712198257447, "epoch": 3.810633946830266, "grad_norm": 0.5078125, "learning_rate": 0.00039189000580845115, "loss": 4.5831, "mean_token_accuracy": 0.2610299110412598, "num_tokens": 75744422.0, "step": 4660 }, { "entropy": 4.772676944732666, "epoch": 3.814723926380368, "grad_norm": 0.474609375, "learning_rate": 0.0003916207359152634, "loss": 4.5457, "mean_token_accuracy": 0.26019563227891923, "num_tokens": 75824402.0, "step": 4665 }, { "entropy": 4.796119689941406, "epoch": 3.81881390593047, "grad_norm": 0.484375, "learning_rate": 0.00039135123744653944, "loss": 4.5768, "mean_token_accuracy": 0.26022201776504517, "num_tokens": 75905998.0, "step": 4670 }, { "entropy": 4.738014888763428, "epoch": 3.8229038854805726, "grad_norm": 0.51171875, "learning_rate": 0.000391081510930494, "loss": 4.4915, "mean_token_accuracy": 0.2647898614406586, "num_tokens": 75987164.0, "step": 4675 }, { "entropy": 4.8066305160522464, "epoch": 3.826993865030675, "grad_norm": 0.4921875, "learning_rate": 0.0003908115568957884, "loss": 4.5925, "mean_token_accuracy": 0.25776982456445696, "num_tokens": 76068182.0, "step": 4680 }, { "entropy": 4.801429319381714, "epoch": 3.8310838445807773, "grad_norm": 0.5, "learning_rate": 0.00039054137587152986, "loss": 4.575, "mean_token_accuracy": 0.2602649927139282, "num_tokens": 76149484.0, "step": 4685 }, { "entropy": 4.7743597507476805, "epoch": 3.8351738241308793, "grad_norm": 0.51953125, "learning_rate": 0.00039027096838727083, "loss": 4.5607, "mean_token_accuracy": 0.26074036955833435, "num_tokens": 76230809.0, "step": 4690 }, { "entropy": 4.833917236328125, "epoch": 3.8392638036809816, "grad_norm": 0.5078125, "learning_rate": 0.0003900003349730073, "loss": 4.6045, "mean_token_accuracy": 0.26039533466100695, "num_tokens": 76311675.0, "step": 4695 }, { "entropy": 4.768689870834351, "epoch": 3.843353783231084, "grad_norm": 0.50390625, "learning_rate": 0.0003897294761591782, "loss": 4.5477, "mean_token_accuracy": 0.26473201960325243, "num_tokens": 76393382.0, "step": 4700 }, { "entropy": 4.8161036491394045, "epoch": 3.847443762781186, "grad_norm": 0.5078125, "learning_rate": 0.0003894583924766642, "loss": 4.5788, "mean_token_accuracy": 0.2625236362218857, "num_tokens": 76475037.0, "step": 4705 }, { "entropy": 4.774548435211182, "epoch": 3.8515337423312883, "grad_norm": 0.494140625, "learning_rate": 0.0003891870844567869, "loss": 4.5398, "mean_token_accuracy": 0.2599763885140419, "num_tokens": 76556551.0, "step": 4710 }, { "entropy": 4.711831712722779, "epoch": 3.8556237218813907, "grad_norm": 0.48046875, "learning_rate": 0.0003889155526313072, "loss": 4.4652, "mean_token_accuracy": 0.2688318133354187, "num_tokens": 76638089.0, "step": 4715 }, { "entropy": 4.785729455947876, "epoch": 3.8597137014314926, "grad_norm": 0.498046875, "learning_rate": 0.000388643797532425, "loss": 4.5442, "mean_token_accuracy": 0.26210551559925077, "num_tokens": 76719686.0, "step": 4720 }, { "entropy": 4.7098531246185305, "epoch": 3.863803680981595, "grad_norm": 0.51171875, "learning_rate": 0.0003883718196927778, "loss": 4.5033, "mean_token_accuracy": 0.2623819649219513, "num_tokens": 76801200.0, "step": 4725 }, { "entropy": 4.7814641952514645, "epoch": 3.8678936605316974, "grad_norm": 0.490234375, "learning_rate": 0.0003880996196454396, "loss": 4.5517, "mean_token_accuracy": 0.2631982520222664, "num_tokens": 76882636.0, "step": 4730 }, { "entropy": 4.770201015472412, "epoch": 3.8719836400817997, "grad_norm": 0.50390625, "learning_rate": 0.00038782719792391975, "loss": 4.5715, "mean_token_accuracy": 0.26133293509483335, "num_tokens": 76963867.0, "step": 4735 }, { "entropy": 4.773094129562378, "epoch": 3.876073619631902, "grad_norm": 0.5, "learning_rate": 0.00038755455506216243, "loss": 4.5301, "mean_token_accuracy": 0.26205867528915405, "num_tokens": 77045497.0, "step": 4740 }, { "entropy": 4.7647865295410154, "epoch": 3.880163599182004, "grad_norm": 0.498046875, "learning_rate": 0.0003872816915945449, "loss": 4.5384, "mean_token_accuracy": 0.26330689489841463, "num_tokens": 77126144.0, "step": 4745 }, { "entropy": 4.741481971740723, "epoch": 3.8842535787321064, "grad_norm": 0.51171875, "learning_rate": 0.000387008608055877, "loss": 4.5434, "mean_token_accuracy": 0.2613819971680641, "num_tokens": 77207217.0, "step": 4750 }, { "entropy": 4.786731815338134, "epoch": 3.888343558282209, "grad_norm": 0.494140625, "learning_rate": 0.00038673530498140003, "loss": 4.5569, "mean_token_accuracy": 0.2603800788521767, "num_tokens": 77288799.0, "step": 4755 }, { "entropy": 4.747005844116211, "epoch": 3.8924335378323107, "grad_norm": 0.4921875, "learning_rate": 0.0003864617829067852, "loss": 4.5339, "mean_token_accuracy": 0.2633078873157501, "num_tokens": 77370355.0, "step": 4760 }, { "entropy": 4.767143726348877, "epoch": 3.896523517382413, "grad_norm": 0.50390625, "learning_rate": 0.0003861880423681334, "loss": 4.5337, "mean_token_accuracy": 0.26211218982934953, "num_tokens": 77451325.0, "step": 4765 }, { "entropy": 4.729104375839233, "epoch": 3.9006134969325155, "grad_norm": 0.486328125, "learning_rate": 0.0003859140839019732, "loss": 4.5262, "mean_token_accuracy": 0.2631507247686386, "num_tokens": 77532922.0, "step": 4770 }, { "entropy": 4.775764465332031, "epoch": 3.9047034764826174, "grad_norm": 0.484375, "learning_rate": 0.0003856399080452609, "loss": 4.5312, "mean_token_accuracy": 0.2637739896774292, "num_tokens": 77614543.0, "step": 4775 }, { "entropy": 4.75821623802185, "epoch": 3.9087934560327198, "grad_norm": 0.484375, "learning_rate": 0.0003853655153353784, "loss": 4.5615, "mean_token_accuracy": 0.26383876949548724, "num_tokens": 77696234.0, "step": 4780 }, { "entropy": 4.780368041992188, "epoch": 3.912883435582822, "grad_norm": 0.51953125, "learning_rate": 0.0003850909063101328, "loss": 4.5373, "mean_token_accuracy": 0.26134357452392576, "num_tokens": 77777478.0, "step": 4785 }, { "entropy": 4.745201587677002, "epoch": 3.916973415132924, "grad_norm": 0.498046875, "learning_rate": 0.0003848160815077553, "loss": 4.5173, "mean_token_accuracy": 0.26424407958984375, "num_tokens": 77857728.0, "step": 4790 }, { "entropy": 4.746929025650024, "epoch": 3.9210633946830264, "grad_norm": 0.51171875, "learning_rate": 0.00038454104146689986, "loss": 4.5387, "mean_token_accuracy": 0.26041311770677567, "num_tokens": 77939134.0, "step": 4795 }, { "entropy": 4.794698667526245, "epoch": 3.925153374233129, "grad_norm": 0.5078125, "learning_rate": 0.0003842657867266423, "loss": 4.5537, "mean_token_accuracy": 0.26144137382507326, "num_tokens": 78020568.0, "step": 4800 }, { "entropy": 4.773353052139282, "epoch": 3.929243353783231, "grad_norm": 0.49609375, "learning_rate": 0.00038399031782647934, "loss": 4.5528, "mean_token_accuracy": 0.26105186641216277, "num_tokens": 78102147.0, "step": 4805 }, { "entropy": 4.770840692520141, "epoch": 3.9333333333333336, "grad_norm": 0.494140625, "learning_rate": 0.0003837146353063275, "loss": 4.5448, "mean_token_accuracy": 0.2626545041799545, "num_tokens": 78183815.0, "step": 4810 }, { "entropy": 4.750923109054566, "epoch": 3.9374233128834355, "grad_norm": 0.5078125, "learning_rate": 0.0003834387397065217, "loss": 4.534, "mean_token_accuracy": 0.264380419254303, "num_tokens": 78265377.0, "step": 4815 }, { "entropy": 4.781521272659302, "epoch": 3.941513292433538, "grad_norm": 0.51171875, "learning_rate": 0.00038316263156781485, "loss": 4.558, "mean_token_accuracy": 0.26068549752235415, "num_tokens": 78347039.0, "step": 4820 }, { "entropy": 4.830083417892456, "epoch": 3.9456032719836402, "grad_norm": 0.50390625, "learning_rate": 0.0003828863114313762, "loss": 4.6217, "mean_token_accuracy": 0.2583851173520088, "num_tokens": 78427791.0, "step": 4825 }, { "entropy": 4.76179986000061, "epoch": 3.949693251533742, "grad_norm": 0.5, "learning_rate": 0.00038260977983879067, "loss": 4.4927, "mean_token_accuracy": 0.26865111887454984, "num_tokens": 78509449.0, "step": 4830 }, { "entropy": 4.727330493927002, "epoch": 3.9537832310838446, "grad_norm": 0.4921875, "learning_rate": 0.0003823330373320575, "loss": 4.5176, "mean_token_accuracy": 0.2647800534963608, "num_tokens": 78590958.0, "step": 4835 }, { "entropy": 4.817589616775512, "epoch": 3.957873210633947, "grad_norm": 0.4765625, "learning_rate": 0.00038205608445358933, "loss": 4.5565, "mean_token_accuracy": 0.26187161207199094, "num_tokens": 78672500.0, "step": 4840 }, { "entropy": 4.742371368408203, "epoch": 3.961963190184049, "grad_norm": 0.494140625, "learning_rate": 0.00038177892174621136, "loss": 4.536, "mean_token_accuracy": 0.2632147118449211, "num_tokens": 78753461.0, "step": 4845 }, { "entropy": 4.820105314254761, "epoch": 3.9660531697341512, "grad_norm": 0.5234375, "learning_rate": 0.00038150154975315974, "loss": 4.5896, "mean_token_accuracy": 0.26224084496498107, "num_tokens": 78834665.0, "step": 4850 }, { "entropy": 4.728599262237549, "epoch": 3.9701431492842536, "grad_norm": 0.4921875, "learning_rate": 0.000381223969018081, "loss": 4.5033, "mean_token_accuracy": 0.26645868718624116, "num_tokens": 78916297.0, "step": 4855 }, { "entropy": 4.740946435928345, "epoch": 3.974233128834356, "grad_norm": 0.51171875, "learning_rate": 0.0003809461800850306, "loss": 4.5141, "mean_token_accuracy": 0.26426759362220764, "num_tokens": 78997453.0, "step": 4860 }, { "entropy": 4.763249301910401, "epoch": 3.9783231083844584, "grad_norm": 0.490234375, "learning_rate": 0.0003806681834984724, "loss": 4.5484, "mean_token_accuracy": 0.26342829763889314, "num_tokens": 79078768.0, "step": 4865 }, { "entropy": 4.76572847366333, "epoch": 3.9824130879345603, "grad_norm": 0.51171875, "learning_rate": 0.00038038997980327694, "loss": 4.5368, "mean_token_accuracy": 0.26370484530925753, "num_tokens": 79159625.0, "step": 4870 }, { "entropy": 4.801229333877563, "epoch": 3.9865030674846627, "grad_norm": 0.51953125, "learning_rate": 0.00038011156954472094, "loss": 4.5817, "mean_token_accuracy": 0.2597121477127075, "num_tokens": 79241054.0, "step": 4875 }, { "entropy": 4.766723918914795, "epoch": 3.990593047034765, "grad_norm": 0.53125, "learning_rate": 0.0003798329532684857, "loss": 4.5591, "mean_token_accuracy": 0.2603618800640106, "num_tokens": 79321912.0, "step": 4880 }, { "entropy": 4.769122743606568, "epoch": 3.994683026584867, "grad_norm": 0.50390625, "learning_rate": 0.00037955413152065666, "loss": 4.5438, "mean_token_accuracy": 0.2623648911714554, "num_tokens": 79402983.0, "step": 4885 }, { "entropy": 4.754130554199219, "epoch": 3.9987730061349693, "grad_norm": 0.486328125, "learning_rate": 0.0003792751048477217, "loss": 4.5, "mean_token_accuracy": 0.2645598530769348, "num_tokens": 79484638.0, "step": 4890 }, { "entropy": 4.827518728044298, "epoch": 4.002453987730061, "grad_norm": 0.51953125, "learning_rate": 0.00037899587379657045, "loss": 4.5054, "mean_token_accuracy": 0.2673109173774719, "num_tokens": 79550949.0, "step": 4895 }, { "entropy": 4.743341112136841, "epoch": 4.006543967280163, "grad_norm": 0.5234375, "learning_rate": 0.0003787164389144932, "loss": 4.4414, "mean_token_accuracy": 0.2697653561830521, "num_tokens": 79632289.0, "step": 4900 }, { "entropy": 4.755812072753907, "epoch": 4.010633946830266, "grad_norm": 0.494140625, "learning_rate": 0.0003784368007491796, "loss": 4.43, "mean_token_accuracy": 0.27117855548858644, "num_tokens": 79713196.0, "step": 4905 }, { "entropy": 4.720384693145752, "epoch": 4.014723926380368, "grad_norm": 0.498046875, "learning_rate": 0.00037815695984871775, "loss": 4.3569, "mean_token_accuracy": 0.2767634868621826, "num_tokens": 79794300.0, "step": 4910 }, { "entropy": 4.702354192733765, "epoch": 4.01881390593047, "grad_norm": 0.50390625, "learning_rate": 0.0003778769167615933, "loss": 4.347, "mean_token_accuracy": 0.2768774747848511, "num_tokens": 79875937.0, "step": 4915 }, { "entropy": 4.761409711837769, "epoch": 4.022903885480573, "grad_norm": 0.5234375, "learning_rate": 0.00037759667203668786, "loss": 4.4063, "mean_token_accuracy": 0.2738190799951553, "num_tokens": 79957550.0, "step": 4920 }, { "entropy": 4.760712909698486, "epoch": 4.026993865030675, "grad_norm": 0.52734375, "learning_rate": 0.00037731622622327865, "loss": 4.3846, "mean_token_accuracy": 0.277262881398201, "num_tokens": 80039135.0, "step": 4925 }, { "entropy": 4.762246465682983, "epoch": 4.0310838445807775, "grad_norm": 0.51171875, "learning_rate": 0.00037703557987103667, "loss": 4.4053, "mean_token_accuracy": 0.2770212650299072, "num_tokens": 80120746.0, "step": 4930 }, { "entropy": 4.782715511322022, "epoch": 4.035173824130879, "grad_norm": 0.53125, "learning_rate": 0.0003767547335300261, "loss": 4.4374, "mean_token_accuracy": 0.2687006890773773, "num_tokens": 80202260.0, "step": 4935 }, { "entropy": 4.789666700363159, "epoch": 4.039263803680981, "grad_norm": 0.51171875, "learning_rate": 0.0003764736877507031, "loss": 4.4396, "mean_token_accuracy": 0.27161316871643065, "num_tokens": 80283826.0, "step": 4940 }, { "entropy": 4.71729736328125, "epoch": 4.043353783231084, "grad_norm": 0.51953125, "learning_rate": 0.0003761924430839148, "loss": 4.3736, "mean_token_accuracy": 0.2782468914985657, "num_tokens": 80364224.0, "step": 4945 }, { "entropy": 4.734089803695679, "epoch": 4.047443762781186, "grad_norm": 0.50390625, "learning_rate": 0.00037591100008089806, "loss": 4.365, "mean_token_accuracy": 0.27749464213848113, "num_tokens": 80445818.0, "step": 4950 }, { "entropy": 4.741703891754151, "epoch": 4.051533742331288, "grad_norm": 0.5078125, "learning_rate": 0.0003756293592932785, "loss": 4.417, "mean_token_accuracy": 0.27450019121170044, "num_tokens": 80526963.0, "step": 4955 }, { "entropy": 4.726750183105469, "epoch": 4.055623721881391, "grad_norm": 0.50390625, "learning_rate": 0.0003753475212730693, "loss": 4.3624, "mean_token_accuracy": 0.2763531029224396, "num_tokens": 80608351.0, "step": 4960 }, { "entropy": 4.7372650623321535, "epoch": 4.059713701431493, "grad_norm": 0.515625, "learning_rate": 0.00037506548657267045, "loss": 4.3921, "mean_token_accuracy": 0.27435546517372134, "num_tokens": 80689492.0, "step": 4965 }, { "entropy": 4.764703321456909, "epoch": 4.063803680981595, "grad_norm": 0.51953125, "learning_rate": 0.0003747832557448672, "loss": 4.4399, "mean_token_accuracy": 0.2734683483839035, "num_tokens": 80771113.0, "step": 4970 }, { "entropy": 4.697296667098999, "epoch": 4.0678936605316975, "grad_norm": 0.51171875, "learning_rate": 0.0003745008293428293, "loss": 4.3556, "mean_token_accuracy": 0.27804059982299806, "num_tokens": 80852368.0, "step": 4975 }, { "entropy": 4.712805986404419, "epoch": 4.0719836400817995, "grad_norm": 0.51171875, "learning_rate": 0.0003742182079201098, "loss": 4.3761, "mean_token_accuracy": 0.2765390634536743, "num_tokens": 80933798.0, "step": 4980 }, { "entropy": 4.720206451416016, "epoch": 4.076073619631902, "grad_norm": 0.515625, "learning_rate": 0.00037393539203064405, "loss": 4.3685, "mean_token_accuracy": 0.2775721073150635, "num_tokens": 81014868.0, "step": 4985 }, { "entropy": 4.727065086364746, "epoch": 4.080163599182004, "grad_norm": 0.515625, "learning_rate": 0.00037365238222874847, "loss": 4.4054, "mean_token_accuracy": 0.27603305876255035, "num_tokens": 81096485.0, "step": 4990 }, { "entropy": 4.666804838180542, "epoch": 4.084253578732106, "grad_norm": 0.50390625, "learning_rate": 0.0003733691790691197, "loss": 4.3018, "mean_token_accuracy": 0.28071958422660825, "num_tokens": 81178104.0, "step": 4995 }, { "entropy": 4.693549489974975, "epoch": 4.088343558282209, "grad_norm": 0.515625, "learning_rate": 0.00037308578310683307, "loss": 4.3697, "mean_token_accuracy": 0.2775652796030045, "num_tokens": 81259342.0, "step": 5000 }, { "epoch": 4.088343558282209, "eval_entropy": 4.554158082800022, "eval_loss": 4.689850330352783, "eval_mean_token_accuracy": 0.2588186653251814, "eval_num_tokens": 81259342.0, "eval_runtime": 65.7584, "eval_samples_per_second": 62.912, "eval_steps_per_second": 7.877, "step": 5000 }, { "entropy": 4.73019552230835, "epoch": 4.092433537832311, "grad_norm": 0.53125, "learning_rate": 0.0003728021948973421, "loss": 4.4132, "mean_token_accuracy": 0.2722706228494644, "num_tokens": 81340319.0, "step": 5005 }, { "entropy": 4.7282239437103275, "epoch": 4.096523517382413, "grad_norm": 0.5234375, "learning_rate": 0.000372518414996477, "loss": 4.3848, "mean_token_accuracy": 0.27392669916152956, "num_tokens": 81421533.0, "step": 5010 }, { "entropy": 4.710113906860352, "epoch": 4.100613496932516, "grad_norm": 0.5078125, "learning_rate": 0.00037223444396044364, "loss": 4.4171, "mean_token_accuracy": 0.2715051531791687, "num_tokens": 81502647.0, "step": 5015 }, { "entropy": 4.80070309638977, "epoch": 4.104703476482618, "grad_norm": 0.5546875, "learning_rate": 0.00037195028234582246, "loss": 4.4696, "mean_token_accuracy": 0.26958937048912046, "num_tokens": 81583856.0, "step": 5020 }, { "entropy": 4.758585453033447, "epoch": 4.1087934560327195, "grad_norm": 0.5234375, "learning_rate": 0.0003716659307095677, "loss": 4.4183, "mean_token_accuracy": 0.2718853384256363, "num_tokens": 81664696.0, "step": 5025 }, { "entropy": 4.736191606521606, "epoch": 4.112883435582822, "grad_norm": 0.515625, "learning_rate": 0.0003713813896090058, "loss": 4.4203, "mean_token_accuracy": 0.27027707695961, "num_tokens": 81745969.0, "step": 5030 }, { "entropy": 4.736065435409546, "epoch": 4.116973415132924, "grad_norm": 0.5078125, "learning_rate": 0.0003710966596018346, "loss": 4.3795, "mean_token_accuracy": 0.27658253610134126, "num_tokens": 81827634.0, "step": 5035 }, { "entropy": 4.728189468383789, "epoch": 4.121063394683026, "grad_norm": 0.5234375, "learning_rate": 0.00037081174124612223, "loss": 4.4171, "mean_token_accuracy": 0.27010842859745027, "num_tokens": 81909159.0, "step": 5040 }, { "entropy": 4.720037841796875, "epoch": 4.125153374233129, "grad_norm": 0.53125, "learning_rate": 0.000370526635100306, "loss": 4.3779, "mean_token_accuracy": 0.27639345824718475, "num_tokens": 81990547.0, "step": 5045 }, { "entropy": 4.7135295391082765, "epoch": 4.129243353783231, "grad_norm": 0.50390625, "learning_rate": 0.00037024134172319113, "loss": 4.4176, "mean_token_accuracy": 0.27186605930328367, "num_tokens": 82072059.0, "step": 5050 }, { "entropy": 4.730686902999878, "epoch": 4.133333333333334, "grad_norm": 0.50390625, "learning_rate": 0.00036995586167395007, "loss": 4.3805, "mean_token_accuracy": 0.27375019490718844, "num_tokens": 82153697.0, "step": 5055 }, { "entropy": 4.734041452407837, "epoch": 4.137423312883436, "grad_norm": 0.515625, "learning_rate": 0.00036967019551212093, "loss": 4.4278, "mean_token_accuracy": 0.27179000079631804, "num_tokens": 82234904.0, "step": 5060 }, { "entropy": 4.68546724319458, "epoch": 4.141513292433538, "grad_norm": 0.5234375, "learning_rate": 0.0003693843437976067, "loss": 4.3415, "mean_token_accuracy": 0.2802614450454712, "num_tokens": 82316023.0, "step": 5065 }, { "entropy": 4.6911194801330565, "epoch": 4.14560327198364, "grad_norm": 0.5, "learning_rate": 0.00036909830709067407, "loss": 4.3938, "mean_token_accuracy": 0.27431837618350985, "num_tokens": 82397048.0, "step": 5070 }, { "entropy": 4.704698514938355, "epoch": 4.149693251533742, "grad_norm": 0.50390625, "learning_rate": 0.00036881208595195214, "loss": 4.3459, "mean_token_accuracy": 0.2787847876548767, "num_tokens": 82478596.0, "step": 5075 }, { "entropy": 4.685911703109741, "epoch": 4.153783231083844, "grad_norm": 0.51171875, "learning_rate": 0.00036852568094243175, "loss": 4.399, "mean_token_accuracy": 0.27456318736076357, "num_tokens": 82560049.0, "step": 5080 }, { "entropy": 4.772367334365844, "epoch": 4.157873210633947, "grad_norm": 0.51953125, "learning_rate": 0.000368239092623464, "loss": 4.4353, "mean_token_accuracy": 0.2711184561252594, "num_tokens": 82640513.0, "step": 5085 }, { "entropy": 4.722985506057739, "epoch": 4.161963190184049, "grad_norm": 0.53125, "learning_rate": 0.0003679523215567593, "loss": 4.4181, "mean_token_accuracy": 0.27188290357589723, "num_tokens": 82721190.0, "step": 5090 }, { "entropy": 4.747667503356934, "epoch": 4.166053169734151, "grad_norm": 0.515625, "learning_rate": 0.00036766536830438605, "loss": 4.3995, "mean_token_accuracy": 0.273905473947525, "num_tokens": 82802500.0, "step": 5095 }, { "entropy": 4.737371969223022, "epoch": 4.170143149284254, "grad_norm": 0.5078125, "learning_rate": 0.0003673782334287701, "loss": 4.4686, "mean_token_accuracy": 0.2665170967578888, "num_tokens": 82884102.0, "step": 5100 }, { "entropy": 4.7536177158355715, "epoch": 4.174233128834356, "grad_norm": 0.51171875, "learning_rate": 0.00036709091749269297, "loss": 4.4346, "mean_token_accuracy": 0.27072409391403196, "num_tokens": 82965160.0, "step": 5105 }, { "entropy": 4.737291097640991, "epoch": 4.178323108384458, "grad_norm": 0.5234375, "learning_rate": 0.00036680342105929124, "loss": 4.4033, "mean_token_accuracy": 0.2732230335474014, "num_tokens": 83046726.0, "step": 5110 }, { "entropy": 4.668428611755371, "epoch": 4.1824130879345605, "grad_norm": 0.5, "learning_rate": 0.0003665157446920551, "loss": 4.3664, "mean_token_accuracy": 0.27558859884738923, "num_tokens": 83127575.0, "step": 5115 }, { "entropy": 4.702426862716675, "epoch": 4.186503067484662, "grad_norm": 0.515625, "learning_rate": 0.0003662278889548275, "loss": 4.3813, "mean_token_accuracy": 0.27629553973674775, "num_tokens": 83209124.0, "step": 5120 }, { "entropy": 4.727795791625977, "epoch": 4.190593047034765, "grad_norm": 0.53125, "learning_rate": 0.0003659398544118031, "loss": 4.3934, "mean_token_accuracy": 0.27651107013225557, "num_tokens": 83290530.0, "step": 5125 }, { "entropy": 4.68690299987793, "epoch": 4.194683026584867, "grad_norm": 0.52734375, "learning_rate": 0.0003656516416275268, "loss": 4.3952, "mean_token_accuracy": 0.2748586177825928, "num_tokens": 83372079.0, "step": 5130 }, { "entropy": 4.77235369682312, "epoch": 4.198773006134969, "grad_norm": 0.5078125, "learning_rate": 0.00036536325116689274, "loss": 4.4307, "mean_token_accuracy": 0.2721753060817719, "num_tokens": 83453729.0, "step": 5135 }, { "entropy": 4.711965942382813, "epoch": 4.202862985685072, "grad_norm": 0.515625, "learning_rate": 0.00036507468359514377, "loss": 4.4063, "mean_token_accuracy": 0.27477738857269285, "num_tokens": 83535392.0, "step": 5140 }, { "entropy": 4.71895022392273, "epoch": 4.206952965235174, "grad_norm": 0.51953125, "learning_rate": 0.0003647859394778695, "loss": 4.3977, "mean_token_accuracy": 0.27320088148117067, "num_tokens": 83616957.0, "step": 5145 }, { "entropy": 4.746178817749024, "epoch": 4.211042944785276, "grad_norm": 0.53125, "learning_rate": 0.0003644970193810056, "loss": 4.4111, "mean_token_accuracy": 0.2707440912723541, "num_tokens": 83698599.0, "step": 5150 }, { "entropy": 4.695309066772461, "epoch": 4.215132924335379, "grad_norm": 0.515625, "learning_rate": 0.0003642079238708327, "loss": 4.3975, "mean_token_accuracy": 0.27356426119804383, "num_tokens": 83780229.0, "step": 5155 }, { "entropy": 4.716104507446289, "epoch": 4.2192229038854805, "grad_norm": 0.53125, "learning_rate": 0.0003639186535139753, "loss": 4.3978, "mean_token_accuracy": 0.27473169267177583, "num_tokens": 83861109.0, "step": 5160 }, { "entropy": 4.682452249526977, "epoch": 4.223312883435582, "grad_norm": 0.5234375, "learning_rate": 0.0003636292088774007, "loss": 4.3638, "mean_token_accuracy": 0.27760191559791564, "num_tokens": 83941796.0, "step": 5165 }, { "entropy": 4.77071533203125, "epoch": 4.227402862985685, "grad_norm": 0.53125, "learning_rate": 0.00036333959052841736, "loss": 4.4628, "mean_token_accuracy": 0.26933829188346864, "num_tokens": 84023400.0, "step": 5170 }, { "entropy": 4.786300802230835, "epoch": 4.231492842535787, "grad_norm": 0.51953125, "learning_rate": 0.00036304979903467474, "loss": 4.4791, "mean_token_accuracy": 0.26967048048973086, "num_tokens": 84104792.0, "step": 5175 }, { "entropy": 4.665366363525391, "epoch": 4.23558282208589, "grad_norm": 0.51953125, "learning_rate": 0.0003627598349641613, "loss": 4.3538, "mean_token_accuracy": 0.2745645225048065, "num_tokens": 84186048.0, "step": 5180 }, { "entropy": 4.746250820159912, "epoch": 4.239672801635992, "grad_norm": 0.5078125, "learning_rate": 0.0003624696988852038, "loss": 4.4173, "mean_token_accuracy": 0.26943006217479704, "num_tokens": 84267216.0, "step": 5185 }, { "entropy": 4.7048177242279055, "epoch": 4.243762781186094, "grad_norm": 0.515625, "learning_rate": 0.0003621793913664663, "loss": 4.4272, "mean_token_accuracy": 0.2724757969379425, "num_tokens": 84348813.0, "step": 5190 }, { "entropy": 4.686121702194214, "epoch": 4.247852760736197, "grad_norm": 0.53125, "learning_rate": 0.00036188891297694886, "loss": 4.3914, "mean_token_accuracy": 0.2723406136035919, "num_tokens": 84430533.0, "step": 5195 }, { "entropy": 4.728496837615967, "epoch": 4.251942740286299, "grad_norm": 0.51953125, "learning_rate": 0.0003615982642859863, "loss": 4.4231, "mean_token_accuracy": 0.2723179399967194, "num_tokens": 84512134.0, "step": 5200 }, { "entropy": 4.681035995483398, "epoch": 4.2560327198364005, "grad_norm": 0.515625, "learning_rate": 0.0003613074458632472, "loss": 4.393, "mean_token_accuracy": 0.27520776093006133, "num_tokens": 84592773.0, "step": 5205 }, { "entropy": 4.727481031417847, "epoch": 4.260122699386503, "grad_norm": 0.51171875, "learning_rate": 0.000361016458278733, "loss": 4.4102, "mean_token_accuracy": 0.2695240557193756, "num_tokens": 84674265.0, "step": 5210 }, { "entropy": 4.704874420166016, "epoch": 4.264212678936605, "grad_norm": 0.50390625, "learning_rate": 0.0003607253021027766, "loss": 4.3963, "mean_token_accuracy": 0.2764166921377182, "num_tokens": 84755851.0, "step": 5215 }, { "entropy": 4.705451011657715, "epoch": 4.268302658486707, "grad_norm": 0.51953125, "learning_rate": 0.00036043397790604126, "loss": 4.419, "mean_token_accuracy": 0.27048486173152925, "num_tokens": 84837443.0, "step": 5220 }, { "entropy": 4.7307366847991945, "epoch": 4.27239263803681, "grad_norm": 0.52734375, "learning_rate": 0.00036014248625951983, "loss": 4.3821, "mean_token_accuracy": 0.27672404050827026, "num_tokens": 84918112.0, "step": 5225 }, { "entropy": 4.71644959449768, "epoch": 4.276482617586912, "grad_norm": 0.515625, "learning_rate": 0.00035985082773453303, "loss": 4.4047, "mean_token_accuracy": 0.27301484644412993, "num_tokens": 84999374.0, "step": 5230 }, { "entropy": 4.722251987457275, "epoch": 4.280572597137015, "grad_norm": 0.51953125, "learning_rate": 0.00035955900290272875, "loss": 4.3829, "mean_token_accuracy": 0.27676635682582856, "num_tokens": 85080984.0, "step": 5235 }, { "entropy": 4.688011598587036, "epoch": 4.284662576687117, "grad_norm": 0.515625, "learning_rate": 0.0003592670123360811, "loss": 4.3897, "mean_token_accuracy": 0.27452185153961184, "num_tokens": 85162629.0, "step": 5240 }, { "entropy": 4.757365131378174, "epoch": 4.288752556237219, "grad_norm": 0.51171875, "learning_rate": 0.00035897485660688863, "loss": 4.4584, "mean_token_accuracy": 0.26980640590190885, "num_tokens": 85243571.0, "step": 5245 }, { "entropy": 4.713522958755493, "epoch": 4.2928425357873214, "grad_norm": 0.5078125, "learning_rate": 0.0003586825362877739, "loss": 4.4238, "mean_token_accuracy": 0.2754086315631866, "num_tokens": 85325217.0, "step": 5250 }, { "entropy": 4.755678796768189, "epoch": 4.296932515337423, "grad_norm": 0.515625, "learning_rate": 0.00035839005195168177, "loss": 4.4383, "mean_token_accuracy": 0.2728460818529129, "num_tokens": 85405925.0, "step": 5255 }, { "entropy": 4.708818101882935, "epoch": 4.301022494887525, "grad_norm": 0.50390625, "learning_rate": 0.0003580974041718791, "loss": 4.3984, "mean_token_accuracy": 0.27643550336360934, "num_tokens": 85487604.0, "step": 5260 }, { "entropy": 4.739409494400024, "epoch": 4.305112474437628, "grad_norm": 0.51953125, "learning_rate": 0.0003578045935219525, "loss": 4.4379, "mean_token_accuracy": 0.27134527564048766, "num_tokens": 85569238.0, "step": 5265 }, { "entropy": 4.722643423080444, "epoch": 4.30920245398773, "grad_norm": 0.51171875, "learning_rate": 0.0003575116205758081, "loss": 4.4157, "mean_token_accuracy": 0.27458052337169647, "num_tokens": 85649880.0, "step": 5270 }, { "entropy": 4.747481012344361, "epoch": 4.313292433537832, "grad_norm": 0.51171875, "learning_rate": 0.0003572184859076702, "loss": 4.4629, "mean_token_accuracy": 0.26937665045261383, "num_tokens": 85731120.0, "step": 5275 }, { "entropy": 4.715378093719482, "epoch": 4.317382413087935, "grad_norm": 0.51953125, "learning_rate": 0.0003569251900920799, "loss": 4.403, "mean_token_accuracy": 0.27391388118267057, "num_tokens": 85812359.0, "step": 5280 }, { "entropy": 4.734446573257446, "epoch": 4.321472392638037, "grad_norm": 0.5234375, "learning_rate": 0.0003566317337038942, "loss": 4.407, "mean_token_accuracy": 0.27438890039920805, "num_tokens": 85893766.0, "step": 5285 }, { "entropy": 4.747154569625854, "epoch": 4.325562372188139, "grad_norm": 0.51171875, "learning_rate": 0.00035633811731828485, "loss": 4.4464, "mean_token_accuracy": 0.2708694994449615, "num_tokens": 85975360.0, "step": 5290 }, { "entropy": 4.7153356075286865, "epoch": 4.3296523517382415, "grad_norm": 0.5390625, "learning_rate": 0.0003560443415107373, "loss": 4.4283, "mean_token_accuracy": 0.2699738711118698, "num_tokens": 86056980.0, "step": 5295 }, { "entropy": 4.708131217956543, "epoch": 4.333742331288343, "grad_norm": 0.52734375, "learning_rate": 0.0003557504068570492, "loss": 4.4014, "mean_token_accuracy": 0.2749925315380096, "num_tokens": 86138062.0, "step": 5300 }, { "entropy": 4.731829357147217, "epoch": 4.337832310838445, "grad_norm": 0.51953125, "learning_rate": 0.00035545631393332967, "loss": 4.434, "mean_token_accuracy": 0.27168336808681487, "num_tokens": 86219326.0, "step": 5305 }, { "entropy": 4.683545398712158, "epoch": 4.341922290388548, "grad_norm": 0.51953125, "learning_rate": 0.00035516206331599813, "loss": 4.3639, "mean_token_accuracy": 0.2776129633188248, "num_tokens": 86299767.0, "step": 5310 }, { "entropy": 4.681959867477417, "epoch": 4.34601226993865, "grad_norm": 0.52734375, "learning_rate": 0.0003548676555817829, "loss": 4.3899, "mean_token_accuracy": 0.2754219532012939, "num_tokens": 86381424.0, "step": 5315 }, { "entropy": 4.668185901641846, "epoch": 4.350102249488753, "grad_norm": 0.5078125, "learning_rate": 0.0003545730913077205, "loss": 4.3758, "mean_token_accuracy": 0.2739670842885971, "num_tokens": 86462823.0, "step": 5320 }, { "entropy": 4.7060998439788815, "epoch": 4.354192229038855, "grad_norm": 0.5078125, "learning_rate": 0.000354278371071154, "loss": 4.4004, "mean_token_accuracy": 0.2745685696601868, "num_tokens": 86544097.0, "step": 5325 }, { "entropy": 4.749334669113159, "epoch": 4.358282208588957, "grad_norm": 0.50390625, "learning_rate": 0.0003539834954497323, "loss": 4.4366, "mean_token_accuracy": 0.2700444906949997, "num_tokens": 86625005.0, "step": 5330 }, { "entropy": 4.7164558410644535, "epoch": 4.36237218813906, "grad_norm": 0.50390625, "learning_rate": 0.0003536884650214087, "loss": 4.4036, "mean_token_accuracy": 0.27285611927509307, "num_tokens": 86706191.0, "step": 5335 }, { "entropy": 4.638006925582886, "epoch": 4.3664621676891615, "grad_norm": 0.51171875, "learning_rate": 0.00035339328036444035, "loss": 4.3581, "mean_token_accuracy": 0.27973835468292235, "num_tokens": 86787844.0, "step": 5340 }, { "entropy": 4.759363031387329, "epoch": 4.370552147239263, "grad_norm": 0.50390625, "learning_rate": 0.0003530979420573861, "loss": 4.4345, "mean_token_accuracy": 0.27088275253772737, "num_tokens": 86868673.0, "step": 5345 }, { "entropy": 4.6891073226928714, "epoch": 4.374642126789366, "grad_norm": 0.51953125, "learning_rate": 0.00035280245067910627, "loss": 4.3735, "mean_token_accuracy": 0.2775958746671677, "num_tokens": 86949880.0, "step": 5350 }, { "entropy": 4.7056732177734375, "epoch": 4.378732106339468, "grad_norm": 0.515625, "learning_rate": 0.0003525068068087613, "loss": 4.3895, "mean_token_accuracy": 0.2754692792892456, "num_tokens": 87031320.0, "step": 5355 }, { "entropy": 4.681880378723145, "epoch": 4.38282208588957, "grad_norm": 0.51953125, "learning_rate": 0.0003522110110258102, "loss": 4.3713, "mean_token_accuracy": 0.2738282591104507, "num_tokens": 87112909.0, "step": 5360 }, { "entropy": 4.687999725341797, "epoch": 4.386912065439673, "grad_norm": 0.515625, "learning_rate": 0.00035191506391001, "loss": 4.4276, "mean_token_accuracy": 0.2707722455263138, "num_tokens": 87193993.0, "step": 5365 }, { "entropy": 4.716944742202759, "epoch": 4.391002044989775, "grad_norm": 0.515625, "learning_rate": 0.0003516189660414142, "loss": 4.4122, "mean_token_accuracy": 0.27415079474449155, "num_tokens": 87275188.0, "step": 5370 }, { "entropy": 4.714711761474609, "epoch": 4.395092024539878, "grad_norm": 0.52734375, "learning_rate": 0.0003513227180003718, "loss": 4.3978, "mean_token_accuracy": 0.27461287975311277, "num_tokens": 87356781.0, "step": 5375 }, { "entropy": 4.759805202484131, "epoch": 4.39918200408998, "grad_norm": 0.51171875, "learning_rate": 0.00035102632036752623, "loss": 4.4389, "mean_token_accuracy": 0.26937533915042877, "num_tokens": 87437984.0, "step": 5380 }, { "entropy": 4.695192432403564, "epoch": 4.4032719836400815, "grad_norm": 0.52734375, "learning_rate": 0.00035072977372381395, "loss": 4.4109, "mean_token_accuracy": 0.27181256413459776, "num_tokens": 87519022.0, "step": 5385 }, { "entropy": 4.78554139137268, "epoch": 4.407361963190184, "grad_norm": 0.50390625, "learning_rate": 0.0003504330786504636, "loss": 4.4481, "mean_token_accuracy": 0.27324004769325255, "num_tokens": 87600415.0, "step": 5390 }, { "entropy": 4.728064250946045, "epoch": 4.411451942740286, "grad_norm": 0.515625, "learning_rate": 0.0003501362357289947, "loss": 4.4282, "mean_token_accuracy": 0.2731639415025711, "num_tokens": 87682142.0, "step": 5395 }, { "entropy": 4.744648361206055, "epoch": 4.415541922290388, "grad_norm": 0.50390625, "learning_rate": 0.00034983924554121665, "loss": 4.4495, "mean_token_accuracy": 0.2723449617624283, "num_tokens": 87762973.0, "step": 5400 }, { "entropy": 4.684806394577026, "epoch": 4.419631901840491, "grad_norm": 0.5078125, "learning_rate": 0.00034954210866922726, "loss": 4.3828, "mean_token_accuracy": 0.2775229841470718, "num_tokens": 87844237.0, "step": 5405 }, { "entropy": 4.702126121520996, "epoch": 4.423721881390593, "grad_norm": 0.50390625, "learning_rate": 0.0003492448256954121, "loss": 4.416, "mean_token_accuracy": 0.27611923813819883, "num_tokens": 87925155.0, "step": 5410 }, { "entropy": 4.708348131179809, "epoch": 4.427811860940695, "grad_norm": 0.5078125, "learning_rate": 0.00034894739720244276, "loss": 4.4028, "mean_token_accuracy": 0.2758689045906067, "num_tokens": 88005893.0, "step": 5415 }, { "entropy": 4.697096681594848, "epoch": 4.431901840490798, "grad_norm": 0.5, "learning_rate": 0.0003486498237732765, "loss": 4.4198, "mean_token_accuracy": 0.2683925271034241, "num_tokens": 88086790.0, "step": 5420 }, { "entropy": 4.7186558723449705, "epoch": 4.4359918200409, "grad_norm": 0.53125, "learning_rate": 0.0003483521059911544, "loss": 4.4186, "mean_token_accuracy": 0.2715620368719101, "num_tokens": 88168171.0, "step": 5425 }, { "entropy": 4.759445905685425, "epoch": 4.4400817995910025, "grad_norm": 0.51953125, "learning_rate": 0.00034805424443960044, "loss": 4.4496, "mean_token_accuracy": 0.2684512048959732, "num_tokens": 88249423.0, "step": 5430 }, { "entropy": 4.708023977279663, "epoch": 4.444171779141104, "grad_norm": 0.5234375, "learning_rate": 0.0003477562397024203, "loss": 4.4106, "mean_token_accuracy": 0.27623399496078493, "num_tokens": 88331004.0, "step": 5435 }, { "entropy": 4.719320631027221, "epoch": 4.448261758691206, "grad_norm": 0.54296875, "learning_rate": 0.0003474580923637007, "loss": 4.4092, "mean_token_accuracy": 0.2749271899461746, "num_tokens": 88412621.0, "step": 5440 }, { "entropy": 4.718741846084595, "epoch": 4.452351738241309, "grad_norm": 0.53125, "learning_rate": 0.0003471598030078074, "loss": 4.3987, "mean_token_accuracy": 0.2753909558057785, "num_tokens": 88494093.0, "step": 5445 }, { "entropy": 4.738852691650391, "epoch": 4.456441717791411, "grad_norm": 0.53515625, "learning_rate": 0.0003468613722193848, "loss": 4.4397, "mean_token_accuracy": 0.27083098888397217, "num_tokens": 88575189.0, "step": 5450 }, { "entropy": 4.731690263748169, "epoch": 4.460531697341513, "grad_norm": 0.51953125, "learning_rate": 0.00034656280058335446, "loss": 4.4433, "mean_token_accuracy": 0.2741995587944984, "num_tokens": 88656767.0, "step": 5455 }, { "entropy": 4.7415751934051515, "epoch": 4.464621676891616, "grad_norm": 0.51171875, "learning_rate": 0.00034626408868491373, "loss": 4.4261, "mean_token_accuracy": 0.27240252792835234, "num_tokens": 88738057.0, "step": 5460 }, { "entropy": 4.728870391845703, "epoch": 4.468711656441718, "grad_norm": 0.51953125, "learning_rate": 0.00034596523710953524, "loss": 4.4362, "mean_token_accuracy": 0.27108282744884493, "num_tokens": 88819060.0, "step": 5465 }, { "entropy": 4.695852041244507, "epoch": 4.47280163599182, "grad_norm": 0.52734375, "learning_rate": 0.00034566624644296527, "loss": 4.3773, "mean_token_accuracy": 0.277285560965538, "num_tokens": 88900530.0, "step": 5470 }, { "entropy": 4.668871593475342, "epoch": 4.4768916155419225, "grad_norm": 0.50390625, "learning_rate": 0.0003453671172712227, "loss": 4.3662, "mean_token_accuracy": 0.276827472448349, "num_tokens": 88981773.0, "step": 5475 }, { "entropy": 4.724203395843506, "epoch": 4.480981595092024, "grad_norm": 0.5234375, "learning_rate": 0.00034506785018059786, "loss": 4.4251, "mean_token_accuracy": 0.27265873551368713, "num_tokens": 89063401.0, "step": 5480 }, { "entropy": 4.68859338760376, "epoch": 4.485071574642127, "grad_norm": 0.51171875, "learning_rate": 0.0003447684457576513, "loss": 4.3791, "mean_token_accuracy": 0.27660117447376253, "num_tokens": 89145030.0, "step": 5485 }, { "entropy": 4.677884197235107, "epoch": 4.489161554192229, "grad_norm": 0.5234375, "learning_rate": 0.000344468904589213, "loss": 4.3738, "mean_token_accuracy": 0.2784330159425735, "num_tokens": 89226686.0, "step": 5490 }, { "entropy": 4.705778789520264, "epoch": 4.493251533742331, "grad_norm": 0.50390625, "learning_rate": 0.0003441692272623807, "loss": 4.4144, "mean_token_accuracy": 0.2744391828775406, "num_tokens": 89308159.0, "step": 5495 }, { "entropy": 4.716770792007447, "epoch": 4.497341513292434, "grad_norm": 0.52734375, "learning_rate": 0.0003438694143645192, "loss": 4.4298, "mean_token_accuracy": 0.27128890752792356, "num_tokens": 89389689.0, "step": 5500 }, { "epoch": 4.497341513292434, "eval_entropy": 4.543432187849951, "eval_loss": 4.665788650512695, "eval_mean_token_accuracy": 0.2611882699847682, "eval_num_tokens": 89389689.0, "eval_runtime": 77.8422, "eval_samples_per_second": 53.146, "eval_steps_per_second": 6.654, "step": 5500 }, { "entropy": 4.711043643951416, "epoch": 4.501431492842536, "grad_norm": 0.52734375, "learning_rate": 0.00034356946648325887, "loss": 4.4339, "mean_token_accuracy": 0.2714413434267044, "num_tokens": 89470394.0, "step": 5505 }, { "entropy": 4.742693042755127, "epoch": 4.505521472392638, "grad_norm": 0.515625, "learning_rate": 0.0003432693842064948, "loss": 4.4235, "mean_token_accuracy": 0.27301117181777956, "num_tokens": 89551381.0, "step": 5510 }, { "entropy": 4.668509101867675, "epoch": 4.509611451942741, "grad_norm": 0.5078125, "learning_rate": 0.00034296916812238525, "loss": 4.3971, "mean_token_accuracy": 0.2757347971200943, "num_tokens": 89632986.0, "step": 5515 }, { "entropy": 4.714461612701416, "epoch": 4.5137014314928425, "grad_norm": 0.5, "learning_rate": 0.000342668818819351, "loss": 4.4033, "mean_token_accuracy": 0.2733671486377716, "num_tokens": 89714645.0, "step": 5520 }, { "entropy": 4.689778900146484, "epoch": 4.5177914110429445, "grad_norm": 0.51953125, "learning_rate": 0.00034236833688607385, "loss": 4.3788, "mean_token_accuracy": 0.2762740761041641, "num_tokens": 89796277.0, "step": 5525 }, { "entropy": 4.697537469863891, "epoch": 4.521881390593047, "grad_norm": 0.51171875, "learning_rate": 0.0003420677229114955, "loss": 4.41, "mean_token_accuracy": 0.27569048702716825, "num_tokens": 89877379.0, "step": 5530 }, { "entropy": 4.7102070331573485, "epoch": 4.525971370143149, "grad_norm": 0.51171875, "learning_rate": 0.0003417669774848165, "loss": 4.4094, "mean_token_accuracy": 0.27565365731716157, "num_tokens": 89958472.0, "step": 5535 }, { "entropy": 4.673035335540772, "epoch": 4.530061349693252, "grad_norm": 0.51953125, "learning_rate": 0.00034146610119549503, "loss": 4.3946, "mean_token_accuracy": 0.2718676865100861, "num_tokens": 90039113.0, "step": 5540 }, { "entropy": 4.723465538024902, "epoch": 4.534151329243354, "grad_norm": 0.515625, "learning_rate": 0.00034116509463324573, "loss": 4.4161, "mean_token_accuracy": 0.26903849840164185, "num_tokens": 90120727.0, "step": 5545 }, { "entropy": 4.729297256469726, "epoch": 4.538241308793456, "grad_norm": 0.515625, "learning_rate": 0.0003408639583880387, "loss": 4.442, "mean_token_accuracy": 0.2692828804254532, "num_tokens": 90202382.0, "step": 5550 }, { "entropy": 4.723666048049926, "epoch": 4.542331288343558, "grad_norm": 0.50390625, "learning_rate": 0.0003405626930500981, "loss": 4.4417, "mean_token_accuracy": 0.2677529126405716, "num_tokens": 90284063.0, "step": 5555 }, { "entropy": 4.703077983856201, "epoch": 4.546421267893661, "grad_norm": 0.5078125, "learning_rate": 0.00034026129920990106, "loss": 4.3741, "mean_token_accuracy": 0.2755285382270813, "num_tokens": 90364922.0, "step": 5560 }, { "entropy": 4.745074367523193, "epoch": 4.550511247443763, "grad_norm": 0.51171875, "learning_rate": 0.0003399597774581766, "loss": 4.4399, "mean_token_accuracy": 0.27432492673397063, "num_tokens": 90446291.0, "step": 5565 }, { "entropy": 4.730029344558716, "epoch": 4.554601226993865, "grad_norm": 0.5078125, "learning_rate": 0.00033965812838590467, "loss": 4.4356, "mean_token_accuracy": 0.2722333788871765, "num_tokens": 90527556.0, "step": 5570 }, { "entropy": 4.715476655960083, "epoch": 4.558691206543967, "grad_norm": 0.515625, "learning_rate": 0.0003393563525843145, "loss": 4.4476, "mean_token_accuracy": 0.2677713245153427, "num_tokens": 90609106.0, "step": 5575 }, { "entropy": 4.728380918502808, "epoch": 4.562781186094069, "grad_norm": 0.51953125, "learning_rate": 0.00033905445064488375, "loss": 4.3917, "mean_token_accuracy": 0.27526160180568693, "num_tokens": 90690269.0, "step": 5580 }, { "entropy": 4.6855494499206545, "epoch": 4.566871165644172, "grad_norm": 0.494140625, "learning_rate": 0.00033875242315933743, "loss": 4.4043, "mean_token_accuracy": 0.2733156532049179, "num_tokens": 90771831.0, "step": 5585 }, { "entropy": 4.714491558074951, "epoch": 4.570961145194274, "grad_norm": 0.51171875, "learning_rate": 0.00033845027071964653, "loss": 4.3844, "mean_token_accuracy": 0.2721282929182053, "num_tokens": 90852988.0, "step": 5590 }, { "entropy": 4.693906259536743, "epoch": 4.575051124744376, "grad_norm": 0.53515625, "learning_rate": 0.000338147993918027, "loss": 4.4093, "mean_token_accuracy": 0.2714533030986786, "num_tokens": 90934085.0, "step": 5595 }, { "entropy": 4.750604581832886, "epoch": 4.579141104294479, "grad_norm": 0.51171875, "learning_rate": 0.0003378455933469384, "loss": 4.4289, "mean_token_accuracy": 0.2718589663505554, "num_tokens": 91015484.0, "step": 5600 }, { "entropy": 4.716768360137939, "epoch": 4.583231083844581, "grad_norm": 0.515625, "learning_rate": 0.000337543069599083, "loss": 4.4233, "mean_token_accuracy": 0.27378277778625487, "num_tokens": 91096680.0, "step": 5605 }, { "entropy": 4.725147151947022, "epoch": 4.587321063394683, "grad_norm": 0.53125, "learning_rate": 0.0003372404232674046, "loss": 4.448, "mean_token_accuracy": 0.2706137031316757, "num_tokens": 91178303.0, "step": 5610 }, { "entropy": 4.722829341888428, "epoch": 4.591411042944785, "grad_norm": 0.52734375, "learning_rate": 0.000336937654945087, "loss": 4.4058, "mean_token_accuracy": 0.27451238930225375, "num_tokens": 91259584.0, "step": 5615 }, { "entropy": 4.656701326370239, "epoch": 4.595501022494887, "grad_norm": 0.515625, "learning_rate": 0.0003366347652255533, "loss": 4.3702, "mean_token_accuracy": 0.27480287849903107, "num_tokens": 91340449.0, "step": 5620 }, { "entropy": 4.682199478149414, "epoch": 4.59959100204499, "grad_norm": 0.5234375, "learning_rate": 0.0003363317547024645, "loss": 4.3849, "mean_token_accuracy": 0.2743862450122833, "num_tokens": 91421611.0, "step": 5625 }, { "entropy": 4.722079229354859, "epoch": 4.603680981595092, "grad_norm": 0.50390625, "learning_rate": 0.0003360286239697184, "loss": 4.4222, "mean_token_accuracy": 0.2695159614086151, "num_tokens": 91503205.0, "step": 5630 }, { "entropy": 4.7281248569488525, "epoch": 4.607770961145194, "grad_norm": 0.51953125, "learning_rate": 0.0003357253736214481, "loss": 4.4463, "mean_token_accuracy": 0.27337758541107177, "num_tokens": 91584153.0, "step": 5635 }, { "entropy": 4.687527704238891, "epoch": 4.611860940695297, "grad_norm": 0.51953125, "learning_rate": 0.00033542200425202174, "loss": 4.4163, "mean_token_accuracy": 0.2738823890686035, "num_tokens": 91665822.0, "step": 5640 }, { "entropy": 4.746928310394287, "epoch": 4.615950920245399, "grad_norm": 0.50390625, "learning_rate": 0.0003351185164560402, "loss": 4.3872, "mean_token_accuracy": 0.2764412611722946, "num_tokens": 91746659.0, "step": 5645 }, { "entropy": 4.715595388412476, "epoch": 4.620040899795501, "grad_norm": 0.515625, "learning_rate": 0.0003348149108283367, "loss": 4.4262, "mean_token_accuracy": 0.27317104637622835, "num_tokens": 91827925.0, "step": 5650 }, { "entropy": 4.698333215713501, "epoch": 4.6241308793456035, "grad_norm": 0.51953125, "learning_rate": 0.0003345111879639755, "loss": 4.4105, "mean_token_accuracy": 0.27452369034290314, "num_tokens": 91909475.0, "step": 5655 }, { "entropy": 4.69792218208313, "epoch": 4.6282208588957054, "grad_norm": 0.51171875, "learning_rate": 0.0003342073484582505, "loss": 4.3968, "mean_token_accuracy": 0.27109030783176424, "num_tokens": 91990989.0, "step": 5660 }, { "entropy": 4.7066840648651125, "epoch": 4.632310838445807, "grad_norm": 0.5234375, "learning_rate": 0.0003339033929066841, "loss": 4.424, "mean_token_accuracy": 0.27283215522766113, "num_tokens": 92072427.0, "step": 5665 }, { "entropy": 4.691405439376831, "epoch": 4.63640081799591, "grad_norm": 0.53125, "learning_rate": 0.00033359932190502647, "loss": 4.3885, "mean_token_accuracy": 0.273253133893013, "num_tokens": 92153182.0, "step": 5670 }, { "entropy": 4.696102905273437, "epoch": 4.640490797546012, "grad_norm": 0.498046875, "learning_rate": 0.0003332951360492538, "loss": 4.4062, "mean_token_accuracy": 0.27294420897960664, "num_tokens": 92234761.0, "step": 5675 }, { "entropy": 4.714269685745239, "epoch": 4.644580777096115, "grad_norm": 0.5, "learning_rate": 0.0003329908359355674, "loss": 4.4267, "mean_token_accuracy": 0.27247470915317534, "num_tokens": 92316326.0, "step": 5680 }, { "entropy": 4.71580286026001, "epoch": 4.648670756646217, "grad_norm": 0.546875, "learning_rate": 0.00033268642216039286, "loss": 4.4346, "mean_token_accuracy": 0.27150200605392455, "num_tokens": 92397475.0, "step": 5685 }, { "entropy": 4.767658996582031, "epoch": 4.652760736196319, "grad_norm": 0.5078125, "learning_rate": 0.00033238189532037803, "loss": 4.4548, "mean_token_accuracy": 0.2686839073896408, "num_tokens": 92479065.0, "step": 5690 }, { "entropy": 4.736369180679321, "epoch": 4.656850715746422, "grad_norm": 0.51171875, "learning_rate": 0.0003320772560123929, "loss": 4.429, "mean_token_accuracy": 0.27476943135261533, "num_tokens": 92560252.0, "step": 5695 }, { "entropy": 4.694738101959229, "epoch": 4.6609406952965236, "grad_norm": 0.51171875, "learning_rate": 0.00033177250483352734, "loss": 4.4037, "mean_token_accuracy": 0.2735146999359131, "num_tokens": 92641797.0, "step": 5700 }, { "entropy": 4.701883602142334, "epoch": 4.6650306748466255, "grad_norm": 0.51171875, "learning_rate": 0.0003314676423810909, "loss": 4.4183, "mean_token_accuracy": 0.27497513592243195, "num_tokens": 92723406.0, "step": 5705 }, { "entropy": 4.707113790512085, "epoch": 4.669120654396728, "grad_norm": 0.53125, "learning_rate": 0.0003311626692526112, "loss": 4.3931, "mean_token_accuracy": 0.2759391188621521, "num_tokens": 92804921.0, "step": 5710 }, { "entropy": 4.733045291900635, "epoch": 4.67321063394683, "grad_norm": 0.5234375, "learning_rate": 0.0003308575860458324, "loss": 4.4443, "mean_token_accuracy": 0.2718596220016479, "num_tokens": 92886548.0, "step": 5715 }, { "entropy": 4.663365745544434, "epoch": 4.677300613496932, "grad_norm": 0.5390625, "learning_rate": 0.000330552393358715, "loss": 4.4011, "mean_token_accuracy": 0.2749068379402161, "num_tokens": 92968162.0, "step": 5720 }, { "entropy": 4.709180641174316, "epoch": 4.681390593047035, "grad_norm": 0.51953125, "learning_rate": 0.0003302470917894335, "loss": 4.403, "mean_token_accuracy": 0.27597300708293915, "num_tokens": 93049851.0, "step": 5725 }, { "entropy": 4.727333402633667, "epoch": 4.685480572597137, "grad_norm": 0.5078125, "learning_rate": 0.0003299416819363763, "loss": 4.4251, "mean_token_accuracy": 0.2702218860387802, "num_tokens": 93131339.0, "step": 5730 }, { "entropy": 4.706945610046387, "epoch": 4.68957055214724, "grad_norm": 0.51171875, "learning_rate": 0.00032963616439814366, "loss": 4.4185, "mean_token_accuracy": 0.2694829210639, "num_tokens": 93212476.0, "step": 5735 }, { "entropy": 4.669017362594604, "epoch": 4.693660531697342, "grad_norm": 0.51171875, "learning_rate": 0.00032933053977354717, "loss": 4.3987, "mean_token_accuracy": 0.27583600878715514, "num_tokens": 93293768.0, "step": 5740 }, { "entropy": 4.716986560821534, "epoch": 4.697750511247444, "grad_norm": 0.52734375, "learning_rate": 0.0003290248086616082, "loss": 4.3969, "mean_token_accuracy": 0.27434879541397095, "num_tokens": 93375428.0, "step": 5745 }, { "entropy": 4.673956441879272, "epoch": 4.7018404907975455, "grad_norm": 0.5078125, "learning_rate": 0.0003287189716615567, "loss": 4.401, "mean_token_accuracy": 0.2732304483652115, "num_tokens": 93457017.0, "step": 5750 }, { "entropy": 4.703333473205566, "epoch": 4.705930470347648, "grad_norm": 0.51171875, "learning_rate": 0.00032841302937283034, "loss": 4.3842, "mean_token_accuracy": 0.2760834485292435, "num_tokens": 93538483.0, "step": 5755 }, { "entropy": 4.710330009460449, "epoch": 4.71002044989775, "grad_norm": 0.50390625, "learning_rate": 0.0003281069823950732, "loss": 4.4224, "mean_token_accuracy": 0.2707503765821457, "num_tokens": 93619247.0, "step": 5760 }, { "entropy": 4.6854576587677, "epoch": 4.714110429447853, "grad_norm": 0.515625, "learning_rate": 0.00032780083132813426, "loss": 4.3927, "mean_token_accuracy": 0.2734291672706604, "num_tokens": 93700771.0, "step": 5765 }, { "entropy": 4.718213415145874, "epoch": 4.718200408997955, "grad_norm": 0.51171875, "learning_rate": 0.00032749457677206696, "loss": 4.45, "mean_token_accuracy": 0.2707763731479645, "num_tokens": 93781972.0, "step": 5770 }, { "entropy": 4.698212480545044, "epoch": 4.722290388548057, "grad_norm": 0.51171875, "learning_rate": 0.00032718821932712706, "loss": 4.3648, "mean_token_accuracy": 0.2789551794528961, "num_tokens": 93863574.0, "step": 5775 }, { "entropy": 4.688614749908448, "epoch": 4.72638036809816, "grad_norm": 0.51953125, "learning_rate": 0.00032688175959377243, "loss": 4.415, "mean_token_accuracy": 0.27334082424640654, "num_tokens": 93944740.0, "step": 5780 }, { "entropy": 4.70204644203186, "epoch": 4.730470347648262, "grad_norm": 0.5078125, "learning_rate": 0.00032657519817266123, "loss": 4.4126, "mean_token_accuracy": 0.27192681431770327, "num_tokens": 94026367.0, "step": 5785 }, { "entropy": 4.703823709487915, "epoch": 4.7345603271983645, "grad_norm": 0.5, "learning_rate": 0.0003262685356646509, "loss": 4.3866, "mean_token_accuracy": 0.27609570920467374, "num_tokens": 94107623.0, "step": 5790 }, { "entropy": 4.712015295028687, "epoch": 4.738650306748466, "grad_norm": 0.52734375, "learning_rate": 0.000325961772670797, "loss": 4.3984, "mean_token_accuracy": 0.27563191056251524, "num_tokens": 94188450.0, "step": 5795 }, { "entropy": 4.695995330810547, "epoch": 4.742740286298568, "grad_norm": 0.48828125, "learning_rate": 0.0003256549097923521, "loss": 4.3899, "mean_token_accuracy": 0.2761635571718216, "num_tokens": 94269769.0, "step": 5800 }, { "entropy": 4.675048398971557, "epoch": 4.74683026584867, "grad_norm": 0.53515625, "learning_rate": 0.00032534794763076463, "loss": 4.3685, "mean_token_accuracy": 0.2758387565612793, "num_tokens": 94350826.0, "step": 5805 }, { "entropy": 4.687421226501465, "epoch": 4.750920245398773, "grad_norm": 0.51953125, "learning_rate": 0.0003250408867876774, "loss": 4.4031, "mean_token_accuracy": 0.2737410306930542, "num_tokens": 94432478.0, "step": 5810 }, { "entropy": 4.7277820110321045, "epoch": 4.755010224948875, "grad_norm": 0.5078125, "learning_rate": 0.0003247337278649268, "loss": 4.4312, "mean_token_accuracy": 0.27192559838294983, "num_tokens": 94513635.0, "step": 5815 }, { "entropy": 4.708493280410766, "epoch": 4.759100204498978, "grad_norm": 0.51171875, "learning_rate": 0.00032442647146454134, "loss": 4.4023, "mean_token_accuracy": 0.2746502161026001, "num_tokens": 94593494.0, "step": 5820 }, { "entropy": 4.684746313095093, "epoch": 4.76319018404908, "grad_norm": 0.5078125, "learning_rate": 0.0003241191181887408, "loss": 4.3673, "mean_token_accuracy": 0.27649718821048735, "num_tokens": 94675131.0, "step": 5825 }, { "entropy": 4.707106113433838, "epoch": 4.767280163599182, "grad_norm": 0.50390625, "learning_rate": 0.00032381166863993456, "loss": 4.4056, "mean_token_accuracy": 0.27251878678798674, "num_tokens": 94756779.0, "step": 5830 }, { "entropy": 4.651648044586182, "epoch": 4.7713701431492845, "grad_norm": 0.51171875, "learning_rate": 0.0003235041234207208, "loss": 4.3727, "mean_token_accuracy": 0.27746806740760804, "num_tokens": 94838384.0, "step": 5835 }, { "entropy": 4.7145240783691404, "epoch": 4.7754601226993865, "grad_norm": 0.515625, "learning_rate": 0.00032319648313388537, "loss": 4.4031, "mean_token_accuracy": 0.27392601370811465, "num_tokens": 94919382.0, "step": 5840 }, { "entropy": 4.727484178543091, "epoch": 4.779550102249488, "grad_norm": 0.5078125, "learning_rate": 0.00032288874838240023, "loss": 4.4503, "mean_token_accuracy": 0.27024946808815004, "num_tokens": 95000549.0, "step": 5845 }, { "entropy": 4.724307203292847, "epoch": 4.783640081799591, "grad_norm": 0.51171875, "learning_rate": 0.0003225809197694225, "loss": 4.4008, "mean_token_accuracy": 0.2741278141736984, "num_tokens": 95082222.0, "step": 5850 }, { "entropy": 4.674021482467651, "epoch": 4.787730061349693, "grad_norm": 0.53125, "learning_rate": 0.0003222729978982935, "loss": 4.3819, "mean_token_accuracy": 0.27415492236614225, "num_tokens": 95163612.0, "step": 5855 }, { "entropy": 4.704116868972778, "epoch": 4.791820040899795, "grad_norm": 0.53515625, "learning_rate": 0.00032196498337253693, "loss": 4.3805, "mean_token_accuracy": 0.2766629010438919, "num_tokens": 95245112.0, "step": 5860 }, { "entropy": 4.683042097091675, "epoch": 4.795910020449898, "grad_norm": 0.53515625, "learning_rate": 0.0003216568767958586, "loss": 4.3609, "mean_token_accuracy": 0.2768434643745422, "num_tokens": 95326574.0, "step": 5865 }, { "entropy": 4.706730556488037, "epoch": 4.8, "grad_norm": 0.51171875, "learning_rate": 0.0003213486787721443, "loss": 4.4246, "mean_token_accuracy": 0.2707619458436966, "num_tokens": 95407651.0, "step": 5870 }, { "entropy": 4.7499689102172855, "epoch": 4.804089979550103, "grad_norm": 0.53125, "learning_rate": 0.00032104038990545935, "loss": 4.4647, "mean_token_accuracy": 0.26709004640579226, "num_tokens": 95488970.0, "step": 5875 }, { "entropy": 4.672450256347656, "epoch": 4.808179959100205, "grad_norm": 0.58984375, "learning_rate": 0.000320732010800047, "loss": 4.3473, "mean_token_accuracy": 0.28012107610702514, "num_tokens": 95570605.0, "step": 5880 }, { "entropy": 4.686946964263916, "epoch": 4.8122699386503065, "grad_norm": 0.52734375, "learning_rate": 0.0003204235420603273, "loss": 4.3965, "mean_token_accuracy": 0.2760658264160156, "num_tokens": 95651851.0, "step": 5885 }, { "entropy": 4.684087419509888, "epoch": 4.816359918200409, "grad_norm": 0.52734375, "learning_rate": 0.00032011498429089616, "loss": 4.3878, "mean_token_accuracy": 0.27788756489753724, "num_tokens": 95733393.0, "step": 5890 }, { "entropy": 4.6571409702301025, "epoch": 4.820449897750511, "grad_norm": 0.515625, "learning_rate": 0.00031980633809652377, "loss": 4.3489, "mean_token_accuracy": 0.2767361462116241, "num_tokens": 95814579.0, "step": 5895 }, { "entropy": 4.699490261077881, "epoch": 4.824539877300613, "grad_norm": 0.51953125, "learning_rate": 0.0003194976040821537, "loss": 4.3865, "mean_token_accuracy": 0.2766541510820389, "num_tokens": 95895497.0, "step": 5900 }, { "entropy": 4.7011901378631595, "epoch": 4.828629856850716, "grad_norm": 0.515625, "learning_rate": 0.00031918878285290173, "loss": 4.3754, "mean_token_accuracy": 0.27975229918956757, "num_tokens": 95977186.0, "step": 5905 }, { "entropy": 4.677326440811157, "epoch": 4.832719836400818, "grad_norm": 0.5234375, "learning_rate": 0.0003188798750140547, "loss": 4.4051, "mean_token_accuracy": 0.27499034702777864, "num_tokens": 96058735.0, "step": 5910 }, { "entropy": 4.690173721313476, "epoch": 4.83680981595092, "grad_norm": 0.51171875, "learning_rate": 0.0003185708811710688, "loss": 4.3593, "mean_token_accuracy": 0.276982182264328, "num_tokens": 96139851.0, "step": 5915 }, { "entropy": 4.662840461730957, "epoch": 4.840899795501023, "grad_norm": 0.51171875, "learning_rate": 0.00031826180192956927, "loss": 4.3697, "mean_token_accuracy": 0.2781296640634537, "num_tokens": 96221427.0, "step": 5920 }, { "entropy": 4.695534706115723, "epoch": 4.844989775051125, "grad_norm": 0.50390625, "learning_rate": 0.00031795263789534834, "loss": 4.4178, "mean_token_accuracy": 0.2717044860124588, "num_tokens": 96303082.0, "step": 5925 }, { "entropy": 4.712454795837402, "epoch": 4.849079754601227, "grad_norm": 0.515625, "learning_rate": 0.00031764338967436455, "loss": 4.4109, "mean_token_accuracy": 0.2756527841091156, "num_tokens": 96384761.0, "step": 5930 }, { "entropy": 4.732760095596314, "epoch": 4.853169734151329, "grad_norm": 0.546875, "learning_rate": 0.0003173340578727416, "loss": 4.4307, "mean_token_accuracy": 0.273481497168541, "num_tokens": 96465879.0, "step": 5935 }, { "entropy": 4.700456523895264, "epoch": 4.857259713701431, "grad_norm": 0.53515625, "learning_rate": 0.0003170246430967669, "loss": 4.3736, "mean_token_accuracy": 0.2790436565876007, "num_tokens": 96546826.0, "step": 5940 }, { "entropy": 4.730433702468872, "epoch": 4.861349693251534, "grad_norm": 0.51171875, "learning_rate": 0.00031671514595289036, "loss": 4.429, "mean_token_accuracy": 0.27219615280628207, "num_tokens": 96627932.0, "step": 5945 }, { "entropy": 4.690513706207275, "epoch": 4.865439672801636, "grad_norm": 0.498046875, "learning_rate": 0.00031640556704772346, "loss": 4.3822, "mean_token_accuracy": 0.27687233686447144, "num_tokens": 96709389.0, "step": 5950 }, { "entropy": 4.658168029785156, "epoch": 4.869529652351738, "grad_norm": 0.51171875, "learning_rate": 0.00031609590698803797, "loss": 4.3624, "mean_token_accuracy": 0.2769273012876511, "num_tokens": 96791013.0, "step": 5955 }, { "entropy": 4.674880170822144, "epoch": 4.873619631901841, "grad_norm": 0.51953125, "learning_rate": 0.00031578616638076466, "loss": 4.3843, "mean_token_accuracy": 0.2747682720422745, "num_tokens": 96872015.0, "step": 5960 }, { "entropy": 4.743522453308105, "epoch": 4.877709611451943, "grad_norm": 0.5234375, "learning_rate": 0.0003154763458329921, "loss": 4.4421, "mean_token_accuracy": 0.2713626593351364, "num_tokens": 96953313.0, "step": 5965 }, { "entropy": 4.7112298011779785, "epoch": 4.881799591002045, "grad_norm": 0.52734375, "learning_rate": 0.0003151664459519658, "loss": 4.4318, "mean_token_accuracy": 0.2723765343427658, "num_tokens": 97034878.0, "step": 5970 }, { "entropy": 4.767510938644409, "epoch": 4.8858895705521475, "grad_norm": 0.51953125, "learning_rate": 0.0003148564673450864, "loss": 4.4679, "mean_token_accuracy": 0.272841015458107, "num_tokens": 97116120.0, "step": 5975 }, { "entropy": 4.657976722717285, "epoch": 4.889979550102249, "grad_norm": 0.50390625, "learning_rate": 0.0003145464106199092, "loss": 4.3788, "mean_token_accuracy": 0.27804608941078185, "num_tokens": 97197694.0, "step": 5980 }, { "entropy": 4.74614315032959, "epoch": 4.894069529652352, "grad_norm": 0.5078125, "learning_rate": 0.0003142362763841423, "loss": 4.4429, "mean_token_accuracy": 0.27136490046977996, "num_tokens": 97279073.0, "step": 5985 }, { "entropy": 4.720199251174927, "epoch": 4.898159509202454, "grad_norm": 0.51171875, "learning_rate": 0.0003139260652456459, "loss": 4.4211, "mean_token_accuracy": 0.2720552355051041, "num_tokens": 97360774.0, "step": 5990 }, { "entropy": 4.705424880981445, "epoch": 4.902249488752556, "grad_norm": 0.5234375, "learning_rate": 0.00031361577781243094, "loss": 4.4195, "mean_token_accuracy": 0.27206000983715056, "num_tokens": 97442401.0, "step": 5995 }, { "entropy": 4.689108657836914, "epoch": 4.906339468302658, "grad_norm": 0.51171875, "learning_rate": 0.00031330541469265776, "loss": 4.3917, "mean_token_accuracy": 0.27495068311691284, "num_tokens": 97523114.0, "step": 6000 }, { "epoch": 4.906339468302658, "eval_entropy": 4.528890131983518, "eval_loss": 4.638745307922363, "eval_mean_token_accuracy": 0.2633246178505043, "eval_num_tokens": 97523114.0, "eval_runtime": 67.5562, "eval_samples_per_second": 61.238, "eval_steps_per_second": 7.668, "step": 6000 }, { "entropy": 4.7361010074615475, "epoch": 4.910429447852761, "grad_norm": 0.51953125, "learning_rate": 0.0003129949764946352, "loss": 4.4507, "mean_token_accuracy": 0.27095101177692416, "num_tokens": 97604381.0, "step": 6005 }, { "entropy": 4.686746311187744, "epoch": 4.914519427402863, "grad_norm": 0.51953125, "learning_rate": 0.0003126844638268192, "loss": 4.3905, "mean_token_accuracy": 0.27397231459617616, "num_tokens": 97685416.0, "step": 6010 }, { "entropy": 4.692208003997803, "epoch": 4.918609406952966, "grad_norm": 0.515625, "learning_rate": 0.0003123738772978115, "loss": 4.4118, "mean_token_accuracy": 0.2731016129255295, "num_tokens": 97767111.0, "step": 6015 }, { "entropy": 4.682928609848022, "epoch": 4.9226993865030675, "grad_norm": 0.51171875, "learning_rate": 0.0003120632175163589, "loss": 4.4162, "mean_token_accuracy": 0.27412803173065187, "num_tokens": 97848813.0, "step": 6020 }, { "entropy": 4.676147747039795, "epoch": 4.926789366053169, "grad_norm": 0.53125, "learning_rate": 0.0003117524850913517, "loss": 4.3871, "mean_token_accuracy": 0.275849723815918, "num_tokens": 97929995.0, "step": 6025 }, { "entropy": 4.701095294952393, "epoch": 4.930879345603272, "grad_norm": 0.515625, "learning_rate": 0.00031144168063182215, "loss": 4.3939, "mean_token_accuracy": 0.2740096032619476, "num_tokens": 98011579.0, "step": 6030 }, { "entropy": 4.748048639297485, "epoch": 4.934969325153374, "grad_norm": 0.51171875, "learning_rate": 0.0003111308047469445, "loss": 4.4512, "mean_token_accuracy": 0.26954575181007384, "num_tokens": 98092218.0, "step": 6035 }, { "entropy": 4.698567295074463, "epoch": 4.939059304703476, "grad_norm": 0.5234375, "learning_rate": 0.0003108198580460322, "loss": 4.4041, "mean_token_accuracy": 0.27610322535037995, "num_tokens": 98173885.0, "step": 6040 }, { "entropy": 4.729394149780274, "epoch": 4.943149284253579, "grad_norm": 0.51953125, "learning_rate": 0.0003105088411385381, "loss": 4.4063, "mean_token_accuracy": 0.27324399948120115, "num_tokens": 98255048.0, "step": 6045 }, { "entropy": 4.671938228607178, "epoch": 4.947239263803681, "grad_norm": 0.52734375, "learning_rate": 0.000310197754634052, "loss": 4.3769, "mean_token_accuracy": 0.2793420970439911, "num_tokens": 98336239.0, "step": 6050 }, { "entropy": 4.690931892395019, "epoch": 4.951329243353783, "grad_norm": 0.51171875, "learning_rate": 0.0003098865991423008, "loss": 4.4039, "mean_token_accuracy": 0.2744638115167618, "num_tokens": 98417233.0, "step": 6055 }, { "entropy": 4.653145790100098, "epoch": 4.955419222903886, "grad_norm": 0.52734375, "learning_rate": 0.0003095753752731463, "loss": 4.3506, "mean_token_accuracy": 0.27893992960453035, "num_tokens": 98498817.0, "step": 6060 }, { "entropy": 4.699347591400146, "epoch": 4.9595092024539875, "grad_norm": 0.515625, "learning_rate": 0.00030926408363658416, "loss": 4.4239, "mean_token_accuracy": 0.2715125888586044, "num_tokens": 98580195.0, "step": 6065 }, { "entropy": 4.7166825294494625, "epoch": 4.96359918200409, "grad_norm": 0.51171875, "learning_rate": 0.0003089527248427432, "loss": 4.4607, "mean_token_accuracy": 0.2713450789451599, "num_tokens": 98660650.0, "step": 6070 }, { "entropy": 4.7218170166015625, "epoch": 4.967689161554192, "grad_norm": 0.52734375, "learning_rate": 0.0003086412995018837, "loss": 4.4268, "mean_token_accuracy": 0.2760044991970062, "num_tokens": 98742296.0, "step": 6075 }, { "entropy": 4.661420726776123, "epoch": 4.971779141104294, "grad_norm": 0.51171875, "learning_rate": 0.00030832980822439624, "loss": 4.3309, "mean_token_accuracy": 0.2822334259748459, "num_tokens": 98823973.0, "step": 6080 }, { "entropy": 4.673013734817505, "epoch": 4.975869120654397, "grad_norm": 0.5078125, "learning_rate": 0.00030801825162080104, "loss": 4.3817, "mean_token_accuracy": 0.2761273801326752, "num_tokens": 98905050.0, "step": 6085 }, { "entropy": 4.709194946289062, "epoch": 4.979959100204499, "grad_norm": 0.52734375, "learning_rate": 0.00030770663030174573, "loss": 4.4312, "mean_token_accuracy": 0.2730088770389557, "num_tokens": 98986669.0, "step": 6090 }, { "entropy": 4.683793830871582, "epoch": 4.984049079754601, "grad_norm": 0.5234375, "learning_rate": 0.0003073949448780053, "loss": 4.3803, "mean_token_accuracy": 0.2775475323200226, "num_tokens": 99068156.0, "step": 6095 }, { "entropy": 4.723601245880127, "epoch": 4.988139059304704, "grad_norm": 0.515625, "learning_rate": 0.0003070831959604803, "loss": 4.4509, "mean_token_accuracy": 0.27361961305141447, "num_tokens": 99149257.0, "step": 6100 }, { "entropy": 4.681419134140015, "epoch": 4.992229038854806, "grad_norm": 0.51171875, "learning_rate": 0.0003067713841601956, "loss": 4.4198, "mean_token_accuracy": 0.2765075981616974, "num_tokens": 99230507.0, "step": 6105 }, { "entropy": 4.703955936431885, "epoch": 4.9963190184049076, "grad_norm": 0.5234375, "learning_rate": 0.0003064595100882993, "loss": 4.4083, "mean_token_accuracy": 0.2746318817138672, "num_tokens": 99311942.0, "step": 6110 }, { "entropy": 4.692366811964247, "epoch": 5.0, "grad_norm": 1.953125, "learning_rate": 0.0003061475743560617, "loss": 4.3375, "mean_token_accuracy": 0.28033769461843705, "num_tokens": 99377525.0, "step": 6115 }, { "entropy": 4.649888277053833, "epoch": 5.004089979550102, "grad_norm": 0.515625, "learning_rate": 0.0003058355775748738, "loss": 4.2452, "mean_token_accuracy": 0.2871495485305786, "num_tokens": 99458255.0, "step": 6120 }, { "entropy": 4.653833818435669, "epoch": 5.008179959100205, "grad_norm": 0.5390625, "learning_rate": 0.00030552352035624646, "loss": 4.2361, "mean_token_accuracy": 0.28646632134914396, "num_tokens": 99539908.0, "step": 6125 }, { "entropy": 4.691452693939209, "epoch": 5.012269938650307, "grad_norm": 0.5234375, "learning_rate": 0.0003052114033118088, "loss": 4.2532, "mean_token_accuracy": 0.28761292397975924, "num_tokens": 99621475.0, "step": 6130 }, { "entropy": 4.6841614723205565, "epoch": 5.016359918200409, "grad_norm": 0.53125, "learning_rate": 0.00030489922705330717, "loss": 4.2591, "mean_token_accuracy": 0.28759315311908723, "num_tokens": 99702873.0, "step": 6135 }, { "entropy": 4.679251623153687, "epoch": 5.020449897750511, "grad_norm": 0.52734375, "learning_rate": 0.00030458699219260425, "loss": 4.2609, "mean_token_accuracy": 0.28631573617458345, "num_tokens": 99783768.0, "step": 6140 }, { "entropy": 4.7359308242797855, "epoch": 5.024539877300613, "grad_norm": 0.53515625, "learning_rate": 0.00030427469934167723, "loss": 4.3275, "mean_token_accuracy": 0.280113485455513, "num_tokens": 99865376.0, "step": 6145 }, { "entropy": 4.7032434940338135, "epoch": 5.028629856850716, "grad_norm": 0.5234375, "learning_rate": 0.00030396234911261736, "loss": 4.2814, "mean_token_accuracy": 0.28740685880184175, "num_tokens": 99946583.0, "step": 6150 }, { "entropy": 4.686667966842651, "epoch": 5.032719836400818, "grad_norm": 0.53515625, "learning_rate": 0.00030364994211762785, "loss": 4.3042, "mean_token_accuracy": 0.28135778307914733, "num_tokens": 100027914.0, "step": 6155 }, { "entropy": 4.700311136245728, "epoch": 5.03680981595092, "grad_norm": 0.546875, "learning_rate": 0.00030333747896902366, "loss": 4.2781, "mean_token_accuracy": 0.2860795557498932, "num_tokens": 100108740.0, "step": 6160 }, { "entropy": 4.658395147323608, "epoch": 5.040899795501023, "grad_norm": 0.515625, "learning_rate": 0.00030302496027922944, "loss": 4.2701, "mean_token_accuracy": 0.2859830528497696, "num_tokens": 100189893.0, "step": 6165 }, { "entropy": 4.684561681747437, "epoch": 5.044989775051125, "grad_norm": 0.53125, "learning_rate": 0.000302712386660779, "loss": 4.2589, "mean_token_accuracy": 0.28538803160190584, "num_tokens": 100271134.0, "step": 6170 }, { "entropy": 4.674528408050537, "epoch": 5.049079754601227, "grad_norm": 0.53125, "learning_rate": 0.00030239975872631356, "loss": 4.2688, "mean_token_accuracy": 0.28839490711689, "num_tokens": 100351888.0, "step": 6175 }, { "entropy": 4.635148620605468, "epoch": 5.0531697341513295, "grad_norm": 0.53125, "learning_rate": 0.000302087077088581, "loss": 4.2422, "mean_token_accuracy": 0.28575087785720826, "num_tokens": 100433408.0, "step": 6180 }, { "entropy": 4.6560217380523685, "epoch": 5.0572597137014315, "grad_norm": 0.53515625, "learning_rate": 0.0003017743423604343, "loss": 4.2988, "mean_token_accuracy": 0.2840681731700897, "num_tokens": 100514402.0, "step": 6185 }, { "entropy": 4.716656112670899, "epoch": 5.061349693251533, "grad_norm": 0.53515625, "learning_rate": 0.00030146155515483057, "loss": 4.318, "mean_token_accuracy": 0.28110859394073484, "num_tokens": 100596050.0, "step": 6190 }, { "entropy": 4.654313564300537, "epoch": 5.065439672801636, "grad_norm": 0.546875, "learning_rate": 0.0003011487160848298, "loss": 4.2356, "mean_token_accuracy": 0.2906017452478409, "num_tokens": 100677625.0, "step": 6195 }, { "entropy": 4.692273044586182, "epoch": 5.069529652351738, "grad_norm": 0.5546875, "learning_rate": 0.0003008358257635935, "loss": 4.2824, "mean_token_accuracy": 0.2859068810939789, "num_tokens": 100759297.0, "step": 6200 }, { "entropy": 4.639877367019653, "epoch": 5.07361963190184, "grad_norm": 0.53515625, "learning_rate": 0.0003005228848043839, "loss": 4.2224, "mean_token_accuracy": 0.2915385693311691, "num_tokens": 100840830.0, "step": 6205 }, { "entropy": 4.679285860061645, "epoch": 5.077709611451943, "grad_norm": 0.55859375, "learning_rate": 0.00030020989382056217, "loss": 4.2819, "mean_token_accuracy": 0.28825623989105226, "num_tokens": 100922418.0, "step": 6210 }, { "entropy": 4.6742747783660885, "epoch": 5.081799591002045, "grad_norm": 0.546875, "learning_rate": 0.0002998968534255877, "loss": 4.2802, "mean_token_accuracy": 0.2836751461029053, "num_tokens": 101004107.0, "step": 6215 }, { "entropy": 4.673584699630737, "epoch": 5.085889570552148, "grad_norm": 0.54296875, "learning_rate": 0.0002995837642330167, "loss": 4.2643, "mean_token_accuracy": 0.2851913541555405, "num_tokens": 101085304.0, "step": 6220 }, { "entropy": 4.676634407043457, "epoch": 5.08997955010225, "grad_norm": 0.53125, "learning_rate": 0.00029927062685650103, "loss": 4.2881, "mean_token_accuracy": 0.2848454713821411, "num_tokens": 101166844.0, "step": 6225 }, { "entropy": 4.672373247146607, "epoch": 5.0940695296523515, "grad_norm": 0.5546875, "learning_rate": 0.00029895744190978706, "loss": 4.2689, "mean_token_accuracy": 0.2853556782007217, "num_tokens": 101247643.0, "step": 6230 }, { "entropy": 4.651356172561646, "epoch": 5.098159509202454, "grad_norm": 0.53515625, "learning_rate": 0.0002986442100067141, "loss": 4.2407, "mean_token_accuracy": 0.2886988341808319, "num_tokens": 101329110.0, "step": 6235 }, { "entropy": 4.680099058151245, "epoch": 5.102249488752556, "grad_norm": 0.55078125, "learning_rate": 0.0002983309317612139, "loss": 4.274, "mean_token_accuracy": 0.29163762331008913, "num_tokens": 101410724.0, "step": 6240 }, { "entropy": 4.6601667404174805, "epoch": 5.106339468302658, "grad_norm": 0.54296875, "learning_rate": 0.0002980176077873087, "loss": 4.2598, "mean_token_accuracy": 0.2862454980611801, "num_tokens": 101491921.0, "step": 6245 }, { "entropy": 4.650085878372193, "epoch": 5.110429447852761, "grad_norm": 0.53125, "learning_rate": 0.0002977042386991107, "loss": 4.2669, "mean_token_accuracy": 0.2843619823455811, "num_tokens": 101573558.0, "step": 6250 }, { "entropy": 4.67042088508606, "epoch": 5.114519427402863, "grad_norm": 0.54296875, "learning_rate": 0.00029739082511082024, "loss": 4.2545, "mean_token_accuracy": 0.28914952278137207, "num_tokens": 101654917.0, "step": 6255 }, { "entropy": 4.6435822486877445, "epoch": 5.118609406952965, "grad_norm": 0.54296875, "learning_rate": 0.0002970773676367249, "loss": 4.2832, "mean_token_accuracy": 0.28309927582740785, "num_tokens": 101736576.0, "step": 6260 }, { "entropy": 4.733069992065429, "epoch": 5.122699386503068, "grad_norm": 0.54296875, "learning_rate": 0.00029676386689119844, "loss": 4.3118, "mean_token_accuracy": 0.2862152338027954, "num_tokens": 101817760.0, "step": 6265 }, { "entropy": 4.64786376953125, "epoch": 5.12678936605317, "grad_norm": 0.5390625, "learning_rate": 0.0002964503234886994, "loss": 4.2527, "mean_token_accuracy": 0.28564445674419403, "num_tokens": 101899060.0, "step": 6270 }, { "entropy": 4.666914939880371, "epoch": 5.130879345603272, "grad_norm": 0.54296875, "learning_rate": 0.0002961367380437698, "loss": 4.2816, "mean_token_accuracy": 0.28614977896213534, "num_tokens": 101980564.0, "step": 6275 }, { "entropy": 4.678023338317871, "epoch": 5.134969325153374, "grad_norm": 0.53515625, "learning_rate": 0.00029582311117103415, "loss": 4.2644, "mean_token_accuracy": 0.28712991178035735, "num_tokens": 102061362.0, "step": 6280 }, { "entropy": 4.638808870315552, "epoch": 5.139059304703476, "grad_norm": 0.55078125, "learning_rate": 0.00029550944348519817, "loss": 4.2455, "mean_token_accuracy": 0.2867669343948364, "num_tokens": 102142718.0, "step": 6285 }, { "entropy": 4.662787866592407, "epoch": 5.143149284253579, "grad_norm": 0.546875, "learning_rate": 0.0002951957356010475, "loss": 4.2618, "mean_token_accuracy": 0.28844820261001586, "num_tokens": 102224275.0, "step": 6290 }, { "entropy": 4.678096866607666, "epoch": 5.147239263803681, "grad_norm": 0.5234375, "learning_rate": 0.00029488198813344663, "loss": 4.2581, "mean_token_accuracy": 0.2875746190547943, "num_tokens": 102305334.0, "step": 6295 }, { "entropy": 4.686196660995483, "epoch": 5.151329243353783, "grad_norm": 0.54296875, "learning_rate": 0.0002945682016973377, "loss": 4.3262, "mean_token_accuracy": 0.2816776394844055, "num_tokens": 102385891.0, "step": 6300 }, { "entropy": 4.654120969772339, "epoch": 5.155419222903886, "grad_norm": 0.5390625, "learning_rate": 0.000294254376907739, "loss": 4.2625, "mean_token_accuracy": 0.28575509786605835, "num_tokens": 102466738.0, "step": 6305 }, { "entropy": 4.681686735153198, "epoch": 5.159509202453988, "grad_norm": 0.53125, "learning_rate": 0.0002939405143797442, "loss": 4.2735, "mean_token_accuracy": 0.2860242992639542, "num_tokens": 102547626.0, "step": 6310 }, { "entropy": 4.637733459472656, "epoch": 5.16359918200409, "grad_norm": 0.546875, "learning_rate": 0.000293626614728521, "loss": 4.2624, "mean_token_accuracy": 0.2862671673297882, "num_tokens": 102629106.0, "step": 6315 }, { "entropy": 4.655986022949219, "epoch": 5.1676891615541924, "grad_norm": 0.546875, "learning_rate": 0.0002933126785693095, "loss": 4.2755, "mean_token_accuracy": 0.28640487790107727, "num_tokens": 102710751.0, "step": 6320 }, { "entropy": 4.727471351623535, "epoch": 5.171779141104294, "grad_norm": 0.53125, "learning_rate": 0.00029299870651742187, "loss": 4.3244, "mean_token_accuracy": 0.28364361822605133, "num_tokens": 102792100.0, "step": 6325 }, { "entropy": 4.5934250831604, "epoch": 5.175869120654396, "grad_norm": 0.53515625, "learning_rate": 0.0002926846991882402, "loss": 4.2092, "mean_token_accuracy": 0.29286899566650393, "num_tokens": 102872651.0, "step": 6330 }, { "entropy": 4.682796859741211, "epoch": 5.179959100204499, "grad_norm": 0.5390625, "learning_rate": 0.0002923706571972159, "loss": 4.3049, "mean_token_accuracy": 0.2806869864463806, "num_tokens": 102954040.0, "step": 6335 }, { "entropy": 4.691317176818847, "epoch": 5.184049079754601, "grad_norm": 0.53125, "learning_rate": 0.00029205658115986823, "loss": 4.2876, "mean_token_accuracy": 0.28517940938472747, "num_tokens": 103035339.0, "step": 6340 }, { "entropy": 4.6680028438568115, "epoch": 5.188139059304704, "grad_norm": 0.546875, "learning_rate": 0.00029174247169178323, "loss": 4.2799, "mean_token_accuracy": 0.28655944764614105, "num_tokens": 103117015.0, "step": 6345 }, { "entropy": 4.646918773651123, "epoch": 5.192229038854806, "grad_norm": 0.5390625, "learning_rate": 0.0002914283294086127, "loss": 4.2882, "mean_token_accuracy": 0.2865586310625076, "num_tokens": 103198602.0, "step": 6350 }, { "entropy": 4.703545951843262, "epoch": 5.196319018404908, "grad_norm": 0.54296875, "learning_rate": 0.00029111415492607233, "loss": 4.2689, "mean_token_accuracy": 0.2874479919672012, "num_tokens": 103279552.0, "step": 6355 }, { "entropy": 4.613406324386597, "epoch": 5.2004089979550105, "grad_norm": 0.53515625, "learning_rate": 0.00029079994885994115, "loss": 4.2479, "mean_token_accuracy": 0.2869574874639511, "num_tokens": 103361027.0, "step": 6360 }, { "entropy": 4.659533739089966, "epoch": 5.2044989775051125, "grad_norm": 0.5390625, "learning_rate": 0.00029048571182606006, "loss": 4.3023, "mean_token_accuracy": 0.28246972560882566, "num_tokens": 103442567.0, "step": 6365 }, { "entropy": 4.674566650390625, "epoch": 5.208588957055214, "grad_norm": 0.55078125, "learning_rate": 0.00029017144444033077, "loss": 4.2745, "mean_token_accuracy": 0.28414817750453947, "num_tokens": 103523720.0, "step": 6370 }, { "entropy": 4.635878086090088, "epoch": 5.212678936605317, "grad_norm": 0.54296875, "learning_rate": 0.0002898571473187143, "loss": 4.2392, "mean_token_accuracy": 0.2866120427846909, "num_tokens": 103605317.0, "step": 6375 }, { "entropy": 4.700038576126099, "epoch": 5.216768916155419, "grad_norm": 0.53515625, "learning_rate": 0.00028954282107722997, "loss": 4.2965, "mean_token_accuracy": 0.2830153852701187, "num_tokens": 103686938.0, "step": 6380 }, { "entropy": 4.67324800491333, "epoch": 5.220858895705521, "grad_norm": 0.546875, "learning_rate": 0.0002892284663319544, "loss": 4.2872, "mean_token_accuracy": 0.2832156181335449, "num_tokens": 103768613.0, "step": 6385 }, { "entropy": 4.627040576934815, "epoch": 5.224948875255624, "grad_norm": 0.5390625, "learning_rate": 0.0002889140836990197, "loss": 4.2558, "mean_token_accuracy": 0.2878110259771347, "num_tokens": 103849713.0, "step": 6390 }, { "entropy": 4.677251672744751, "epoch": 5.229038854805726, "grad_norm": 0.53125, "learning_rate": 0.00028859967379461294, "loss": 4.3013, "mean_token_accuracy": 0.28213165402412416, "num_tokens": 103931397.0, "step": 6395 }, { "entropy": 4.6230072498321535, "epoch": 5.233128834355828, "grad_norm": 0.546875, "learning_rate": 0.0002882852372349746, "loss": 4.2515, "mean_token_accuracy": 0.28757057189941404, "num_tokens": 104012657.0, "step": 6400 }, { "entropy": 4.692079067230225, "epoch": 5.237218813905931, "grad_norm": 0.54296875, "learning_rate": 0.0002879707746363973, "loss": 4.2933, "mean_token_accuracy": 0.28521048426628115, "num_tokens": 104094047.0, "step": 6405 }, { "entropy": 4.6593328475952145, "epoch": 5.2413087934560325, "grad_norm": 0.52734375, "learning_rate": 0.00028765628661522484, "loss": 4.2602, "mean_token_accuracy": 0.2895059138536453, "num_tokens": 104175251.0, "step": 6410 }, { "entropy": 4.649277448654175, "epoch": 5.245398773006135, "grad_norm": 0.53515625, "learning_rate": 0.00028734177378785057, "loss": 4.2718, "mean_token_accuracy": 0.2875450402498245, "num_tokens": 104256662.0, "step": 6415 }, { "entropy": 4.720139360427856, "epoch": 5.249488752556237, "grad_norm": 0.5625, "learning_rate": 0.00028702723677071683, "loss": 4.3228, "mean_token_accuracy": 0.2819401353597641, "num_tokens": 104338300.0, "step": 6420 }, { "entropy": 4.684942960739136, "epoch": 5.253578732106339, "grad_norm": 0.5625, "learning_rate": 0.0002867126761803131, "loss": 4.289, "mean_token_accuracy": 0.28458996713161466, "num_tokens": 104419899.0, "step": 6425 }, { "entropy": 4.689294528961182, "epoch": 5.257668711656442, "grad_norm": 0.53515625, "learning_rate": 0.00028639809263317516, "loss": 4.3064, "mean_token_accuracy": 0.27976694107055666, "num_tokens": 104501535.0, "step": 6430 }, { "entropy": 4.638687801361084, "epoch": 5.261758691206544, "grad_norm": 0.546875, "learning_rate": 0.0002860834867458838, "loss": 4.2672, "mean_token_accuracy": 0.2854257315397263, "num_tokens": 104582232.0, "step": 6435 }, { "entropy": 4.706684160232544, "epoch": 5.265848670756646, "grad_norm": 0.54296875, "learning_rate": 0.0002857688591350636, "loss": 4.3141, "mean_token_accuracy": 0.282974636554718, "num_tokens": 104663884.0, "step": 6440 }, { "entropy": 4.67949686050415, "epoch": 5.269938650306749, "grad_norm": 0.5625, "learning_rate": 0.0002854542104173816, "loss": 4.2719, "mean_token_accuracy": 0.28465049564838407, "num_tokens": 104745493.0, "step": 6445 }, { "entropy": 4.640626001358032, "epoch": 5.274028629856851, "grad_norm": 0.53125, "learning_rate": 0.00028513954120954637, "loss": 4.2369, "mean_token_accuracy": 0.28675337731838224, "num_tokens": 104827070.0, "step": 6450 }, { "entropy": 4.679956769943237, "epoch": 5.2781186094069525, "grad_norm": 0.546875, "learning_rate": 0.0002848248521283066, "loss": 4.302, "mean_token_accuracy": 0.282174751162529, "num_tokens": 104908365.0, "step": 6455 }, { "entropy": 4.710936546325684, "epoch": 5.282208588957055, "grad_norm": 0.53515625, "learning_rate": 0.00028451014379044984, "loss": 4.3235, "mean_token_accuracy": 0.2818226605653763, "num_tokens": 104989188.0, "step": 6460 }, { "entropy": 4.690206861495971, "epoch": 5.286298568507157, "grad_norm": 0.55078125, "learning_rate": 0.0002841954168128014, "loss": 4.3304, "mean_token_accuracy": 0.2803535670042038, "num_tokens": 105070767.0, "step": 6465 }, { "entropy": 4.645661878585815, "epoch": 5.29038854805726, "grad_norm": 0.54296875, "learning_rate": 0.00028388067181222334, "loss": 4.242, "mean_token_accuracy": 0.2888016074895859, "num_tokens": 105152418.0, "step": 6470 }, { "entropy": 4.657642841339111, "epoch": 5.294478527607362, "grad_norm": 0.53515625, "learning_rate": 0.00028356590940561277, "loss": 4.272, "mean_token_accuracy": 0.28808585107326506, "num_tokens": 105233727.0, "step": 6475 }, { "entropy": 4.681158208847046, "epoch": 5.298568507157464, "grad_norm": 0.5390625, "learning_rate": 0.000283251130209901, "loss": 4.3122, "mean_token_accuracy": 0.2822310596704483, "num_tokens": 105314938.0, "step": 6480 }, { "entropy": 4.632437753677368, "epoch": 5.302658486707567, "grad_norm": 0.5625, "learning_rate": 0.0002829363348420523, "loss": 4.2641, "mean_token_accuracy": 0.28190656900405886, "num_tokens": 105396683.0, "step": 6485 }, { "entropy": 4.648112440109253, "epoch": 5.306748466257669, "grad_norm": 0.52734375, "learning_rate": 0.0002826215239190626, "loss": 4.2656, "mean_token_accuracy": 0.28479851484298707, "num_tokens": 105477376.0, "step": 6490 }, { "entropy": 4.691006231307983, "epoch": 5.310838445807771, "grad_norm": 0.53515625, "learning_rate": 0.0002823066980579583, "loss": 4.3079, "mean_token_accuracy": 0.2849561393260956, "num_tokens": 105558948.0, "step": 6495 }, { "entropy": 4.622942304611206, "epoch": 5.3149284253578735, "grad_norm": 0.52734375, "learning_rate": 0.0002819918578757952, "loss": 4.2492, "mean_token_accuracy": 0.2880774766206741, "num_tokens": 105640622.0, "step": 6500 }, { "epoch": 5.3149284253578735, "eval_entropy": 4.5095378441239875, "eval_loss": 4.627156734466553, "eval_mean_token_accuracy": 0.26533692039578116, "eval_num_tokens": 105640622.0, "eval_runtime": 66.9755, "eval_samples_per_second": 61.769, "eval_steps_per_second": 7.734, "step": 6500 }, { "entropy": 4.678410768508911, "epoch": 5.319018404907975, "grad_norm": 0.54296875, "learning_rate": 0.0002816770039896569, "loss": 4.2775, "mean_token_accuracy": 0.28632205426692964, "num_tokens": 105721134.0, "step": 6505 }, { "entropy": 4.6475409984588625, "epoch": 5.323108384458077, "grad_norm": 0.54296875, "learning_rate": 0.00028136213701665416, "loss": 4.2551, "mean_token_accuracy": 0.2866376847028732, "num_tokens": 105802379.0, "step": 6510 }, { "entropy": 4.673610544204712, "epoch": 5.32719836400818, "grad_norm": 0.53125, "learning_rate": 0.00028104725757392316, "loss": 4.2923, "mean_token_accuracy": 0.28542805910110475, "num_tokens": 105883531.0, "step": 6515 }, { "entropy": 4.696327018737793, "epoch": 5.331288343558282, "grad_norm": 0.55859375, "learning_rate": 0.0002807323662786248, "loss": 4.3272, "mean_token_accuracy": 0.2813657850027084, "num_tokens": 105965139.0, "step": 6520 }, { "entropy": 4.698910140991211, "epoch": 5.335378323108385, "grad_norm": 0.5859375, "learning_rate": 0.00028041746374794286, "loss": 4.3296, "mean_token_accuracy": 0.2815339595079422, "num_tokens": 106046447.0, "step": 6525 }, { "entropy": 4.648953771591186, "epoch": 5.339468302658487, "grad_norm": 0.546875, "learning_rate": 0.0002801025505990834, "loss": 4.2592, "mean_token_accuracy": 0.2858387380838394, "num_tokens": 106127694.0, "step": 6530 }, { "entropy": 4.683491277694702, "epoch": 5.343558282208589, "grad_norm": 0.5703125, "learning_rate": 0.0002797876274492732, "loss": 4.269, "mean_token_accuracy": 0.28837797939777376, "num_tokens": 106208922.0, "step": 6535 }, { "entropy": 4.6552589416503904, "epoch": 5.347648261758692, "grad_norm": 0.54296875, "learning_rate": 0.0002794726949157585, "loss": 4.3031, "mean_token_accuracy": 0.28137973248958587, "num_tokens": 106289645.0, "step": 6540 }, { "entropy": 4.724133539199829, "epoch": 5.3517382413087935, "grad_norm": 0.52734375, "learning_rate": 0.00027915775361580424, "loss": 4.3206, "mean_token_accuracy": 0.28649517595767976, "num_tokens": 106370851.0, "step": 6545 }, { "entropy": 4.639323377609253, "epoch": 5.355828220858895, "grad_norm": 0.55078125, "learning_rate": 0.0002788428041666924, "loss": 4.2838, "mean_token_accuracy": 0.2850149214267731, "num_tokens": 106452577.0, "step": 6550 }, { "entropy": 4.662995958328247, "epoch": 5.359918200408998, "grad_norm": 0.55078125, "learning_rate": 0.00027852784718572076, "loss": 4.306, "mean_token_accuracy": 0.2830405205488205, "num_tokens": 106534290.0, "step": 6555 }, { "entropy": 4.647454500198364, "epoch": 5.3640081799591, "grad_norm": 0.5546875, "learning_rate": 0.00027821288329020217, "loss": 4.2704, "mean_token_accuracy": 0.2870162814855576, "num_tokens": 106615460.0, "step": 6560 }, { "entropy": 4.684663534164429, "epoch": 5.368098159509202, "grad_norm": 0.54296875, "learning_rate": 0.00027789791309746275, "loss": 4.3074, "mean_token_accuracy": 0.2831536322832108, "num_tokens": 106696537.0, "step": 6565 }, { "entropy": 4.677006101608276, "epoch": 5.372188139059305, "grad_norm": 0.5390625, "learning_rate": 0.00027758293722484126, "loss": 4.3049, "mean_token_accuracy": 0.28370130360126494, "num_tokens": 106777663.0, "step": 6570 }, { "entropy": 4.656659555435181, "epoch": 5.376278118609407, "grad_norm": 0.55078125, "learning_rate": 0.0002772679562896872, "loss": 4.2446, "mean_token_accuracy": 0.2909713417291641, "num_tokens": 106858898.0, "step": 6575 }, { "entropy": 4.666804695129395, "epoch": 5.38036809815951, "grad_norm": 0.54296875, "learning_rate": 0.0002769529709093604, "loss": 4.31, "mean_token_accuracy": 0.2837449699640274, "num_tokens": 106940498.0, "step": 6580 }, { "entropy": 4.679633331298828, "epoch": 5.384458077709612, "grad_norm": 0.54296875, "learning_rate": 0.00027663798170122914, "loss": 4.2899, "mean_token_accuracy": 0.2848484665155411, "num_tokens": 107022121.0, "step": 6585 }, { "entropy": 4.626410627365113, "epoch": 5.3885480572597135, "grad_norm": 0.5390625, "learning_rate": 0.00027632298928266925, "loss": 4.2652, "mean_token_accuracy": 0.2820416182279587, "num_tokens": 107103803.0, "step": 6590 }, { "entropy": 4.63424162864685, "epoch": 5.392638036809816, "grad_norm": 0.53125, "learning_rate": 0.0002760079942710629, "loss": 4.2523, "mean_token_accuracy": 0.28843666017055514, "num_tokens": 107185206.0, "step": 6595 }, { "entropy": 4.669785499572754, "epoch": 5.396728016359918, "grad_norm": 0.5390625, "learning_rate": 0.00027569299728379733, "loss": 4.2786, "mean_token_accuracy": 0.28459797203540804, "num_tokens": 107266769.0, "step": 6600 }, { "entropy": 4.6584704399108885, "epoch": 5.40081799591002, "grad_norm": 0.55859375, "learning_rate": 0.00027537799893826375, "loss": 4.2813, "mean_token_accuracy": 0.2869125932455063, "num_tokens": 107347524.0, "step": 6605 }, { "entropy": 4.6850261211395265, "epoch": 5.404907975460123, "grad_norm": 0.54296875, "learning_rate": 0.00027506299985185573, "loss": 4.2872, "mean_token_accuracy": 0.28456632792949677, "num_tokens": 107429209.0, "step": 6610 }, { "entropy": 4.626144981384277, "epoch": 5.408997955010225, "grad_norm": 0.53125, "learning_rate": 0.00027474800064196863, "loss": 4.2601, "mean_token_accuracy": 0.2875842183828354, "num_tokens": 107509871.0, "step": 6615 }, { "entropy": 4.691272163391114, "epoch": 5.413087934560327, "grad_norm": 0.55859375, "learning_rate": 0.00027443300192599797, "loss": 4.3265, "mean_token_accuracy": 0.2830471247434616, "num_tokens": 107590696.0, "step": 6620 }, { "entropy": 4.679490852355957, "epoch": 5.41717791411043, "grad_norm": 0.5390625, "learning_rate": 0.0002741180043213381, "loss": 4.3155, "mean_token_accuracy": 0.2820233255624771, "num_tokens": 107672302.0, "step": 6625 }, { "entropy": 4.660007810592651, "epoch": 5.421267893660532, "grad_norm": 0.546875, "learning_rate": 0.00027380300844538136, "loss": 4.2621, "mean_token_accuracy": 0.2853416293859482, "num_tokens": 107753825.0, "step": 6630 }, { "entropy": 4.598485374450684, "epoch": 5.425357873210634, "grad_norm": 0.53515625, "learning_rate": 0.00027348801491551675, "loss": 4.2418, "mean_token_accuracy": 0.2878331959247589, "num_tokens": 107835008.0, "step": 6635 }, { "entropy": 4.672269201278686, "epoch": 5.429447852760736, "grad_norm": 0.53125, "learning_rate": 0.00027317302434912854, "loss": 4.2916, "mean_token_accuracy": 0.2842607319355011, "num_tokens": 107916413.0, "step": 6640 }, { "entropy": 4.680228233337402, "epoch": 5.433537832310838, "grad_norm": 0.546875, "learning_rate": 0.00027285803736359525, "loss": 4.302, "mean_token_accuracy": 0.28488736748695376, "num_tokens": 107997934.0, "step": 6645 }, { "entropy": 4.66015887260437, "epoch": 5.43762781186094, "grad_norm": 0.546875, "learning_rate": 0.0002725430545762885, "loss": 4.2762, "mean_token_accuracy": 0.28541135787963867, "num_tokens": 108079615.0, "step": 6650 }, { "entropy": 4.669615745544434, "epoch": 5.441717791411043, "grad_norm": 0.546875, "learning_rate": 0.0002722280766045714, "loss": 4.2907, "mean_token_accuracy": 0.28098976612091064, "num_tokens": 108160970.0, "step": 6655 }, { "entropy": 4.648915195465088, "epoch": 5.445807770961145, "grad_norm": 0.5234375, "learning_rate": 0.0002719131040657978, "loss": 4.2719, "mean_token_accuracy": 0.284726682305336, "num_tokens": 108242117.0, "step": 6660 }, { "entropy": 4.625553274154663, "epoch": 5.449897750511248, "grad_norm": 0.54296875, "learning_rate": 0.00027159813757731105, "loss": 4.2509, "mean_token_accuracy": 0.2873865783214569, "num_tokens": 108323658.0, "step": 6665 }, { "entropy": 4.664066314697266, "epoch": 5.45398773006135, "grad_norm": 0.53125, "learning_rate": 0.00027128317775644236, "loss": 4.291, "mean_token_accuracy": 0.283258056640625, "num_tokens": 108405238.0, "step": 6670 }, { "entropy": 4.634313488006592, "epoch": 5.458077709611452, "grad_norm": 0.54296875, "learning_rate": 0.00027096822522050995, "loss": 4.2756, "mean_token_accuracy": 0.28346610367298125, "num_tokens": 108486864.0, "step": 6675 }, { "entropy": 4.64615306854248, "epoch": 5.4621676891615545, "grad_norm": 0.52734375, "learning_rate": 0.0002706532805868179, "loss": 4.2733, "mean_token_accuracy": 0.28266108632087705, "num_tokens": 108568146.0, "step": 6680 }, { "entropy": 4.686797618865967, "epoch": 5.466257668711656, "grad_norm": 0.55078125, "learning_rate": 0.00027033834447265473, "loss": 4.302, "mean_token_accuracy": 0.28264990746974944, "num_tokens": 108649800.0, "step": 6685 }, { "entropy": 4.686423444747925, "epoch": 5.470347648261758, "grad_norm": 0.54296875, "learning_rate": 0.000270023417495292, "loss": 4.3518, "mean_token_accuracy": 0.278118959069252, "num_tokens": 108730533.0, "step": 6690 }, { "entropy": 4.660273551940918, "epoch": 5.474437627811861, "grad_norm": 0.53125, "learning_rate": 0.00026970850027198395, "loss": 4.2809, "mean_token_accuracy": 0.28778275549411775, "num_tokens": 108812187.0, "step": 6695 }, { "entropy": 4.718528509140015, "epoch": 5.478527607361963, "grad_norm": 0.55078125, "learning_rate": 0.00026939359341996506, "loss": 4.3353, "mean_token_accuracy": 0.28079943656921386, "num_tokens": 108893850.0, "step": 6700 }, { "entropy": 4.647251415252685, "epoch": 5.482617586912065, "grad_norm": 0.5390625, "learning_rate": 0.0002690786975564498, "loss": 4.2708, "mean_token_accuracy": 0.28663626611232756, "num_tokens": 108975365.0, "step": 6705 }, { "entropy": 4.67765440940857, "epoch": 5.486707566462168, "grad_norm": 0.54296875, "learning_rate": 0.0002687638132986314, "loss": 4.3052, "mean_token_accuracy": 0.2841372311115265, "num_tokens": 109056850.0, "step": 6710 }, { "entropy": 4.638742351531983, "epoch": 5.49079754601227, "grad_norm": 0.54296875, "learning_rate": 0.0002684489412636796, "loss": 4.2513, "mean_token_accuracy": 0.29138111472129824, "num_tokens": 109138211.0, "step": 6715 }, { "entropy": 4.65220365524292, "epoch": 5.494887525562373, "grad_norm": 0.546875, "learning_rate": 0.0002681340820687408, "loss": 4.2762, "mean_token_accuracy": 0.28447321355342864, "num_tokens": 109219703.0, "step": 6720 }, { "entropy": 4.667575168609619, "epoch": 5.4989775051124745, "grad_norm": 0.53125, "learning_rate": 0.00026781923633093603, "loss": 4.2895, "mean_token_accuracy": 0.283936607837677, "num_tokens": 109300694.0, "step": 6725 }, { "entropy": 4.633440637588501, "epoch": 5.5030674846625764, "grad_norm": 0.55859375, "learning_rate": 0.0002675044046673599, "loss": 4.2475, "mean_token_accuracy": 0.2883361101150513, "num_tokens": 109381952.0, "step": 6730 }, { "entropy": 4.698451232910156, "epoch": 5.507157464212679, "grad_norm": 0.55078125, "learning_rate": 0.0002671895876950795, "loss": 4.3122, "mean_token_accuracy": 0.2892479121685028, "num_tokens": 109463155.0, "step": 6735 }, { "entropy": 4.669783973693848, "epoch": 5.511247443762781, "grad_norm": 0.5390625, "learning_rate": 0.00026687478603113325, "loss": 4.2817, "mean_token_accuracy": 0.2856166958808899, "num_tokens": 109544240.0, "step": 6740 }, { "entropy": 4.718588924407959, "epoch": 5.515337423312883, "grad_norm": 0.54296875, "learning_rate": 0.00026656000029252907, "loss": 4.3455, "mean_token_accuracy": 0.2799066811800003, "num_tokens": 109625275.0, "step": 6745 }, { "entropy": 4.697926712036133, "epoch": 5.519427402862986, "grad_norm": 0.54296875, "learning_rate": 0.0002662452310962445, "loss": 4.3319, "mean_token_accuracy": 0.2839725911617279, "num_tokens": 109706083.0, "step": 6750 }, { "entropy": 4.711754703521729, "epoch": 5.523517382413088, "grad_norm": 0.55078125, "learning_rate": 0.00026593047905922376, "loss": 4.3397, "mean_token_accuracy": 0.2816936582326889, "num_tokens": 109787589.0, "step": 6755 }, { "entropy": 4.6670488834381105, "epoch": 5.52760736196319, "grad_norm": 0.54296875, "learning_rate": 0.000265615744798378, "loss": 4.308, "mean_token_accuracy": 0.28136477768421175, "num_tokens": 109868761.0, "step": 6760 }, { "entropy": 4.643719816207886, "epoch": 5.531697341513293, "grad_norm": 0.53515625, "learning_rate": 0.00026530102893058353, "loss": 4.2656, "mean_token_accuracy": 0.2869348257780075, "num_tokens": 109950486.0, "step": 6765 }, { "entropy": 4.641464710235596, "epoch": 5.5357873210633946, "grad_norm": 0.5390625, "learning_rate": 0.00026498633207268027, "loss": 4.2847, "mean_token_accuracy": 0.2822311490774155, "num_tokens": 110031718.0, "step": 6770 }, { "entropy": 4.697618865966797, "epoch": 5.539877300613497, "grad_norm": 0.5546875, "learning_rate": 0.0002646716548414712, "loss": 4.327, "mean_token_accuracy": 0.2798221081495285, "num_tokens": 110112786.0, "step": 6775 }, { "entropy": 4.713689136505127, "epoch": 5.543967280163599, "grad_norm": 0.5625, "learning_rate": 0.00026435699785372065, "loss": 4.3307, "mean_token_accuracy": 0.283614507317543, "num_tokens": 110193970.0, "step": 6780 }, { "entropy": 4.6377281665802, "epoch": 5.548057259713701, "grad_norm": 0.54296875, "learning_rate": 0.0002640423617261533, "loss": 4.2852, "mean_token_accuracy": 0.2818080306053162, "num_tokens": 110274471.0, "step": 6785 }, { "entropy": 4.662336540222168, "epoch": 5.552147239263804, "grad_norm": 0.53515625, "learning_rate": 0.000263727747075453, "loss": 4.2914, "mean_token_accuracy": 0.28411161303520205, "num_tokens": 110355497.0, "step": 6790 }, { "entropy": 4.638145589828492, "epoch": 5.556237218813906, "grad_norm": 0.51953125, "learning_rate": 0.00026341315451826157, "loss": 4.2431, "mean_token_accuracy": 0.2901862859725952, "num_tokens": 110436544.0, "step": 6795 }, { "entropy": 4.6620666027069095, "epoch": 5.560327198364008, "grad_norm": 0.53125, "learning_rate": 0.0002630985846711773, "loss": 4.2822, "mean_token_accuracy": 0.2851398915052414, "num_tokens": 110517940.0, "step": 6800 }, { "entropy": 4.657286977767944, "epoch": 5.564417177914111, "grad_norm": 0.5234375, "learning_rate": 0.00026278403815075416, "loss": 4.2912, "mean_token_accuracy": 0.28130720257759095, "num_tokens": 110599215.0, "step": 6805 }, { "entropy": 4.643634939193726, "epoch": 5.568507157464213, "grad_norm": 0.546875, "learning_rate": 0.0002624695155735004, "loss": 4.26, "mean_token_accuracy": 0.28722981214523313, "num_tokens": 110680642.0, "step": 6810 }, { "entropy": 4.641807174682617, "epoch": 5.572597137014315, "grad_norm": 0.52734375, "learning_rate": 0.0002621550175558772, "loss": 4.2689, "mean_token_accuracy": 0.285606449842453, "num_tokens": 110762191.0, "step": 6815 }, { "entropy": 4.610067749023438, "epoch": 5.576687116564417, "grad_norm": 0.53515625, "learning_rate": 0.00026184054471429776, "loss": 4.2521, "mean_token_accuracy": 0.2848559945821762, "num_tokens": 110843892.0, "step": 6820 }, { "entropy": 4.646746063232422, "epoch": 5.580777096114519, "grad_norm": 0.53515625, "learning_rate": 0.00026152609766512587, "loss": 4.2794, "mean_token_accuracy": 0.28496140241622925, "num_tokens": 110925154.0, "step": 6825 }, { "entropy": 4.688165903091431, "epoch": 5.584867075664622, "grad_norm": 0.54296875, "learning_rate": 0.0002612116770246747, "loss": 4.3382, "mean_token_accuracy": 0.27987136542797086, "num_tokens": 111006467.0, "step": 6830 }, { "entropy": 4.696931076049805, "epoch": 5.588957055214724, "grad_norm": 0.55859375, "learning_rate": 0.0002608972834092057, "loss": 4.3286, "mean_token_accuracy": 0.2851775884628296, "num_tokens": 111087828.0, "step": 6835 }, { "entropy": 4.663779354095459, "epoch": 5.593047034764826, "grad_norm": 0.55859375, "learning_rate": 0.0002605829174349276, "loss": 4.2818, "mean_token_accuracy": 0.284160190820694, "num_tokens": 111168921.0, "step": 6840 }, { "entropy": 4.673430681228638, "epoch": 5.597137014314928, "grad_norm": 0.55078125, "learning_rate": 0.00026026857971799433, "loss": 4.292, "mean_token_accuracy": 0.28558131158351896, "num_tokens": 111250231.0, "step": 6845 }, { "entropy": 4.671381330490112, "epoch": 5.601226993865031, "grad_norm": 0.546875, "learning_rate": 0.00025995427087450524, "loss": 4.2867, "mean_token_accuracy": 0.2867391496896744, "num_tokens": 111331545.0, "step": 6850 }, { "entropy": 4.643342065811157, "epoch": 5.605316973415133, "grad_norm": 0.5390625, "learning_rate": 0.00025963999152050246, "loss": 4.2763, "mean_token_accuracy": 0.28680709600448606, "num_tokens": 111413199.0, "step": 6855 }, { "entropy": 4.6324633121490475, "epoch": 5.6094069529652355, "grad_norm": 0.53125, "learning_rate": 0.0002593257422719705, "loss": 4.2447, "mean_token_accuracy": 0.28889880180358884, "num_tokens": 111494823.0, "step": 6860 }, { "entropy": 4.629706525802613, "epoch": 5.613496932515337, "grad_norm": 0.55078125, "learning_rate": 0.0002590115237448351, "loss": 4.2281, "mean_token_accuracy": 0.2891985595226288, "num_tokens": 111576207.0, "step": 6865 }, { "entropy": 4.6416624069213865, "epoch": 5.617586912065439, "grad_norm": 0.52734375, "learning_rate": 0.00025869733655496137, "loss": 4.2717, "mean_token_accuracy": 0.28554926812648773, "num_tokens": 111657968.0, "step": 6870 }, { "entropy": 4.637938213348389, "epoch": 5.621676891615542, "grad_norm": 0.546875, "learning_rate": 0.00025838318131815336, "loss": 4.2675, "mean_token_accuracy": 0.2868941843509674, "num_tokens": 111738725.0, "step": 6875 }, { "entropy": 4.618788528442383, "epoch": 5.625766871165644, "grad_norm": 0.53125, "learning_rate": 0.0002580690586501524, "loss": 4.2267, "mean_token_accuracy": 0.2907662779092789, "num_tokens": 111819305.0, "step": 6880 }, { "entropy": 4.651620244979858, "epoch": 5.629856850715746, "grad_norm": 0.53125, "learning_rate": 0.00025775496916663585, "loss": 4.2966, "mean_token_accuracy": 0.2856370806694031, "num_tokens": 111900977.0, "step": 6885 }, { "entropy": 4.632920742034912, "epoch": 5.633946830265849, "grad_norm": 0.52734375, "learning_rate": 0.00025744091348321605, "loss": 4.2461, "mean_token_accuracy": 0.2898207813501358, "num_tokens": 111982643.0, "step": 6890 }, { "entropy": 4.675538015365601, "epoch": 5.638036809815951, "grad_norm": 0.54296875, "learning_rate": 0.0002571268922154393, "loss": 4.2913, "mean_token_accuracy": 0.28617195785045624, "num_tokens": 112063786.0, "step": 6895 }, { "entropy": 4.666571378707886, "epoch": 5.642126789366053, "grad_norm": 0.53515625, "learning_rate": 0.0002568129059787842, "loss": 4.3099, "mean_token_accuracy": 0.28237518966197966, "num_tokens": 112144284.0, "step": 6900 }, { "entropy": 4.629270792007446, "epoch": 5.6462167689161555, "grad_norm": 0.53125, "learning_rate": 0.0002564989553886609, "loss": 4.2478, "mean_token_accuracy": 0.2867598235607147, "num_tokens": 112225010.0, "step": 6905 }, { "entropy": 4.66056342124939, "epoch": 5.6503067484662575, "grad_norm": 0.5390625, "learning_rate": 0.00025618504106040955, "loss": 4.3111, "mean_token_accuracy": 0.28435586392879486, "num_tokens": 112306294.0, "step": 6910 }, { "entropy": 4.64202470779419, "epoch": 5.65439672801636, "grad_norm": 0.546875, "learning_rate": 0.0002558711636092992, "loss": 4.2672, "mean_token_accuracy": 0.2884962946176529, "num_tokens": 112387884.0, "step": 6915 }, { "entropy": 4.6340094089508055, "epoch": 5.658486707566462, "grad_norm": 0.51953125, "learning_rate": 0.0002555573236505267, "loss": 4.2699, "mean_token_accuracy": 0.28879443407058714, "num_tokens": 112469309.0, "step": 6920 }, { "entropy": 4.667138910293579, "epoch": 5.662576687116564, "grad_norm": 0.546875, "learning_rate": 0.0002552435217992154, "loss": 4.2791, "mean_token_accuracy": 0.28828377425670626, "num_tokens": 112550881.0, "step": 6925 }, { "entropy": 4.668040132522583, "epoch": 5.666666666666667, "grad_norm": 0.55078125, "learning_rate": 0.00025492975867041397, "loss": 4.3054, "mean_token_accuracy": 0.2852639019489288, "num_tokens": 112632590.0, "step": 6930 }, { "entropy": 4.698174905776978, "epoch": 5.670756646216769, "grad_norm": 0.5546875, "learning_rate": 0.00025461603487909504, "loss": 4.3331, "mean_token_accuracy": 0.28268762230873107, "num_tokens": 112713956.0, "step": 6935 }, { "entropy": 4.641023588180542, "epoch": 5.674846625766871, "grad_norm": 0.55078125, "learning_rate": 0.0002543023510401545, "loss": 4.2846, "mean_token_accuracy": 0.2842786341905594, "num_tokens": 112794943.0, "step": 6940 }, { "entropy": 4.698601913452149, "epoch": 5.678936605316974, "grad_norm": 0.546875, "learning_rate": 0.0002539887077684095, "loss": 4.3287, "mean_token_accuracy": 0.27716316282749176, "num_tokens": 112876570.0, "step": 6945 }, { "entropy": 4.650067138671875, "epoch": 5.683026584867076, "grad_norm": 0.54296875, "learning_rate": 0.0002536751056785979, "loss": 4.2772, "mean_token_accuracy": 0.2844658404588699, "num_tokens": 112958253.0, "step": 6950 }, { "entropy": 4.597990417480469, "epoch": 5.6871165644171775, "grad_norm": 0.546875, "learning_rate": 0.00025336154538537696, "loss": 4.2213, "mean_token_accuracy": 0.291637995839119, "num_tokens": 113038972.0, "step": 6955 }, { "entropy": 4.666133165359497, "epoch": 5.69120654396728, "grad_norm": 0.53515625, "learning_rate": 0.00025304802750332174, "loss": 4.2848, "mean_token_accuracy": 0.2853614926338196, "num_tokens": 113120234.0, "step": 6960 }, { "entropy": 4.658086442947388, "epoch": 5.695296523517382, "grad_norm": 0.54296875, "learning_rate": 0.0002527345526469245, "loss": 4.2881, "mean_token_accuracy": 0.28501429557800295, "num_tokens": 113200722.0, "step": 6965 }, { "entropy": 4.662501430511474, "epoch": 5.699386503067485, "grad_norm": 0.53125, "learning_rate": 0.0002524211214305926, "loss": 4.308, "mean_token_accuracy": 0.2821039229631424, "num_tokens": 113282358.0, "step": 6970 }, { "entropy": 4.682625389099121, "epoch": 5.703476482617587, "grad_norm": 0.5390625, "learning_rate": 0.00025210773446864866, "loss": 4.2906, "mean_token_accuracy": 0.28561550080776216, "num_tokens": 113363950.0, "step": 6975 }, { "entropy": 4.648300647735596, "epoch": 5.707566462167689, "grad_norm": 0.54296875, "learning_rate": 0.000251794392375328, "loss": 4.2826, "mean_token_accuracy": 0.2862802863121033, "num_tokens": 113445547.0, "step": 6980 }, { "entropy": 4.690504169464111, "epoch": 5.711656441717792, "grad_norm": 0.55859375, "learning_rate": 0.000251481095764778, "loss": 4.3509, "mean_token_accuracy": 0.2786434471607208, "num_tokens": 113526953.0, "step": 6985 }, { "entropy": 4.69014048576355, "epoch": 5.715746421267894, "grad_norm": 0.53125, "learning_rate": 0.00025116784525105725, "loss": 4.3306, "mean_token_accuracy": 0.28160480558872225, "num_tokens": 113608120.0, "step": 6990 }, { "entropy": 4.645735836029052, "epoch": 5.719836400817996, "grad_norm": 0.53125, "learning_rate": 0.00025085464144813377, "loss": 4.2846, "mean_token_accuracy": 0.2855375647544861, "num_tokens": 113689797.0, "step": 6995 }, { "entropy": 4.689887523651123, "epoch": 5.723926380368098, "grad_norm": 0.5390625, "learning_rate": 0.00025054148496988387, "loss": 4.3051, "mean_token_accuracy": 0.28934896290302276, "num_tokens": 113771242.0, "step": 7000 }, { "epoch": 5.723926380368098, "eval_entropy": 4.502966878957269, "eval_loss": 4.608926296234131, "eval_mean_token_accuracy": 0.2670684251143205, "eval_num_tokens": 113771242.0, "eval_runtime": 66.4392, "eval_samples_per_second": 62.267, "eval_steps_per_second": 7.797, "step": 7000 }, { "entropy": 4.6424377918243405, "epoch": 5.7280163599182, "grad_norm": 0.53515625, "learning_rate": 0.00025022837643009143, "loss": 4.2905, "mean_token_accuracy": 0.284592804312706, "num_tokens": 113851741.0, "step": 7005 }, { "entropy": 4.640987682342529, "epoch": 5.732106339468302, "grad_norm": 0.53515625, "learning_rate": 0.0002499153164424464, "loss": 4.2776, "mean_token_accuracy": 0.282452318072319, "num_tokens": 113932881.0, "step": 7010 }, { "entropy": 4.677769899368286, "epoch": 5.736196319018405, "grad_norm": 0.54296875, "learning_rate": 0.00024960230562054284, "loss": 4.2894, "mean_token_accuracy": 0.28492253720760347, "num_tokens": 114014444.0, "step": 7015 }, { "entropy": 4.673404884338379, "epoch": 5.740286298568507, "grad_norm": 0.5390625, "learning_rate": 0.00024928934457787955, "loss": 4.2935, "mean_token_accuracy": 0.28641400337219236, "num_tokens": 114095238.0, "step": 7020 }, { "entropy": 4.672257757186889, "epoch": 5.74437627811861, "grad_norm": 0.53125, "learning_rate": 0.00024897643392785695, "loss": 4.2841, "mean_token_accuracy": 0.28467186987400056, "num_tokens": 114176692.0, "step": 7025 }, { "entropy": 4.632853746414185, "epoch": 5.748466257668712, "grad_norm": 0.5546875, "learning_rate": 0.0002486635742837769, "loss": 4.2599, "mean_token_accuracy": 0.28704401552677156, "num_tokens": 114258127.0, "step": 7030 }, { "entropy": 4.695837688446045, "epoch": 5.752556237218814, "grad_norm": 0.546875, "learning_rate": 0.0002483507662588414, "loss": 4.2996, "mean_token_accuracy": 0.2860493063926697, "num_tokens": 114339721.0, "step": 7035 }, { "entropy": 4.668770790100098, "epoch": 5.7566462167689165, "grad_norm": 0.53515625, "learning_rate": 0.00024803801046615134, "loss": 4.3287, "mean_token_accuracy": 0.2811815798282623, "num_tokens": 114421189.0, "step": 7040 }, { "entropy": 4.655260801315308, "epoch": 5.7607361963190185, "grad_norm": 0.5234375, "learning_rate": 0.0002477253075187048, "loss": 4.3157, "mean_token_accuracy": 0.2774512320756912, "num_tokens": 114502721.0, "step": 7045 }, { "entropy": 4.654748916625977, "epoch": 5.76482617586912, "grad_norm": 0.5546875, "learning_rate": 0.0002474126580293969, "loss": 4.243, "mean_token_accuracy": 0.28693695962429044, "num_tokens": 114584095.0, "step": 7050 }, { "entropy": 4.625263929367065, "epoch": 5.768916155419223, "grad_norm": 0.52734375, "learning_rate": 0.0002471000626110176, "loss": 4.2727, "mean_token_accuracy": 0.2822159260511398, "num_tokens": 114665777.0, "step": 7055 }, { "entropy": 4.671080303192139, "epoch": 5.773006134969325, "grad_norm": 0.55078125, "learning_rate": 0.00024678752187625076, "loss": 4.2803, "mean_token_accuracy": 0.285338369011879, "num_tokens": 114746810.0, "step": 7060 }, { "entropy": 4.640094137191772, "epoch": 5.777096114519427, "grad_norm": 0.53125, "learning_rate": 0.0002464750364376735, "loss": 4.286, "mean_token_accuracy": 0.2817827224731445, "num_tokens": 114828538.0, "step": 7065 }, { "entropy": 4.6721940517425535, "epoch": 5.78118609406953, "grad_norm": 0.53515625, "learning_rate": 0.00024616260690775416, "loss": 4.2919, "mean_token_accuracy": 0.28390816748142245, "num_tokens": 114909812.0, "step": 7070 }, { "entropy": 4.652301597595215, "epoch": 5.785276073619632, "grad_norm": 0.55859375, "learning_rate": 0.00024585023389885177, "loss": 4.2828, "mean_token_accuracy": 0.2851872980594635, "num_tokens": 114991006.0, "step": 7075 }, { "entropy": 4.676189947128296, "epoch": 5.789366053169735, "grad_norm": 0.55078125, "learning_rate": 0.00024553791802321435, "loss": 4.2993, "mean_token_accuracy": 0.28512121438980104, "num_tokens": 115072624.0, "step": 7080 }, { "entropy": 4.655922698974609, "epoch": 5.793456032719837, "grad_norm": 0.53125, "learning_rate": 0.00024522565989297805, "loss": 4.2719, "mean_token_accuracy": 0.2858207762241364, "num_tokens": 115154300.0, "step": 7085 }, { "entropy": 4.6232789039611815, "epoch": 5.7975460122699385, "grad_norm": 0.546875, "learning_rate": 0.00024491346012016596, "loss": 4.2683, "mean_token_accuracy": 0.28604949116706846, "num_tokens": 115235272.0, "step": 7090 }, { "entropy": 4.671456241607666, "epoch": 5.80163599182004, "grad_norm": 0.5390625, "learning_rate": 0.00024460131931668664, "loss": 4.3089, "mean_token_accuracy": 0.2833879321813583, "num_tokens": 115316607.0, "step": 7095 }, { "entropy": 4.631782341003418, "epoch": 5.805725971370143, "grad_norm": 0.53515625, "learning_rate": 0.000244289238094333, "loss": 4.2745, "mean_token_accuracy": 0.28584389984607694, "num_tokens": 115398197.0, "step": 7100 }, { "entropy": 4.637339019775391, "epoch": 5.809815950920245, "grad_norm": 0.54296875, "learning_rate": 0.00024397721706478136, "loss": 4.2701, "mean_token_accuracy": 0.28333486914634703, "num_tokens": 115479379.0, "step": 7105 }, { "entropy": 4.658144140243531, "epoch": 5.813905930470348, "grad_norm": 0.5390625, "learning_rate": 0.00024366525683959007, "loss": 4.2639, "mean_token_accuracy": 0.2861873239278793, "num_tokens": 115560135.0, "step": 7110 }, { "entropy": 4.6280097484588625, "epoch": 5.81799591002045, "grad_norm": 0.546875, "learning_rate": 0.000243353358030198, "loss": 4.2749, "mean_token_accuracy": 0.2864055663347244, "num_tokens": 115641583.0, "step": 7115 }, { "entropy": 4.620970630645752, "epoch": 5.822085889570552, "grad_norm": 0.53515625, "learning_rate": 0.000243041521247924, "loss": 4.2386, "mean_token_accuracy": 0.29097036719322206, "num_tokens": 115722750.0, "step": 7120 }, { "entropy": 4.601324796676636, "epoch": 5.826175869120655, "grad_norm": 0.546875, "learning_rate": 0.0002427297471039652, "loss": 4.2141, "mean_token_accuracy": 0.29036099910736085, "num_tokens": 115804365.0, "step": 7125 }, { "entropy": 4.647102165222168, "epoch": 5.830265848670757, "grad_norm": 0.53515625, "learning_rate": 0.0002424180362093959, "loss": 4.284, "mean_token_accuracy": 0.2843719571828842, "num_tokens": 115885544.0, "step": 7130 }, { "entropy": 4.658620834350586, "epoch": 5.8343558282208585, "grad_norm": 0.5390625, "learning_rate": 0.00024210638917516647, "loss": 4.291, "mean_token_accuracy": 0.2831098258495331, "num_tokens": 115966849.0, "step": 7135 }, { "entropy": 4.710108852386474, "epoch": 5.838445807770961, "grad_norm": 0.55078125, "learning_rate": 0.00024179480661210234, "loss": 4.3449, "mean_token_accuracy": 0.28017795979976656, "num_tokens": 116048171.0, "step": 7140 }, { "entropy": 4.675672435760498, "epoch": 5.842535787321063, "grad_norm": 0.53125, "learning_rate": 0.00024148328913090218, "loss": 4.3116, "mean_token_accuracy": 0.280153751373291, "num_tokens": 116129647.0, "step": 7145 }, { "entropy": 4.64854621887207, "epoch": 5.846625766871165, "grad_norm": 0.55078125, "learning_rate": 0.00024117183734213726, "loss": 4.2878, "mean_token_accuracy": 0.28648664653301237, "num_tokens": 116210985.0, "step": 7150 }, { "entropy": 4.643886137008667, "epoch": 5.850715746421268, "grad_norm": 0.53125, "learning_rate": 0.00024086045185625044, "loss": 4.2678, "mean_token_accuracy": 0.28948957920074464, "num_tokens": 116292548.0, "step": 7155 }, { "entropy": 4.659030246734619, "epoch": 5.85480572597137, "grad_norm": 0.53125, "learning_rate": 0.00024054913328355393, "loss": 4.2766, "mean_token_accuracy": 0.28766130805015566, "num_tokens": 116373959.0, "step": 7160 }, { "entropy": 4.6368225574493405, "epoch": 5.858895705521473, "grad_norm": 0.53125, "learning_rate": 0.0002402378822342295, "loss": 4.2532, "mean_token_accuracy": 0.2867351919412613, "num_tokens": 116455535.0, "step": 7165 }, { "entropy": 4.655530834197998, "epoch": 5.862985685071575, "grad_norm": 0.55078125, "learning_rate": 0.00023992669931832594, "loss": 4.2886, "mean_token_accuracy": 0.28377875685691833, "num_tokens": 116536627.0, "step": 7170 }, { "entropy": 4.6410479068756105, "epoch": 5.867075664621677, "grad_norm": 0.54296875, "learning_rate": 0.0002396155851457591, "loss": 4.2683, "mean_token_accuracy": 0.2879430949687958, "num_tokens": 116618247.0, "step": 7175 }, { "entropy": 4.663385009765625, "epoch": 5.871165644171779, "grad_norm": 0.5390625, "learning_rate": 0.0002393045403263097, "loss": 4.3237, "mean_token_accuracy": 0.2816352307796478, "num_tokens": 116699807.0, "step": 7180 }, { "entropy": 4.667257976531983, "epoch": 5.875255623721881, "grad_norm": 0.5390625, "learning_rate": 0.00023899356546962258, "loss": 4.3108, "mean_token_accuracy": 0.2852670520544052, "num_tokens": 116781080.0, "step": 7185 }, { "entropy": 4.679271602630616, "epoch": 5.879345603271983, "grad_norm": 0.5390625, "learning_rate": 0.00023868266118520553, "loss": 4.3117, "mean_token_accuracy": 0.2820241391658783, "num_tokens": 116862680.0, "step": 7190 }, { "entropy": 4.679209566116333, "epoch": 5.883435582822086, "grad_norm": 0.55859375, "learning_rate": 0.00023837182808242812, "loss": 4.3117, "mean_token_accuracy": 0.282609298825264, "num_tokens": 116944310.0, "step": 7195 }, { "entropy": 4.648283243179321, "epoch": 5.887525562372188, "grad_norm": 0.53515625, "learning_rate": 0.00023806106677052007, "loss": 4.2589, "mean_token_accuracy": 0.2879555135965347, "num_tokens": 117025417.0, "step": 7200 }, { "entropy": 4.655107116699218, "epoch": 5.89161554192229, "grad_norm": 0.53515625, "learning_rate": 0.00023775037785857075, "loss": 4.3034, "mean_token_accuracy": 0.2838689684867859, "num_tokens": 117106168.0, "step": 7205 }, { "entropy": 4.710268306732178, "epoch": 5.895705521472393, "grad_norm": 0.546875, "learning_rate": 0.00023743976195552752, "loss": 4.3399, "mean_token_accuracy": 0.2859070271253586, "num_tokens": 117187628.0, "step": 7210 }, { "entropy": 4.656140279769898, "epoch": 5.899795501022495, "grad_norm": 0.53515625, "learning_rate": 0.00023712921967019446, "loss": 4.2777, "mean_token_accuracy": 0.288422504067421, "num_tokens": 117268597.0, "step": 7215 }, { "entropy": 4.656024694442749, "epoch": 5.9038854805725975, "grad_norm": 0.546875, "learning_rate": 0.0002368187516112316, "loss": 4.27, "mean_token_accuracy": 0.2874049633741379, "num_tokens": 117349439.0, "step": 7220 }, { "entropy": 4.590969228744507, "epoch": 5.9079754601226995, "grad_norm": 0.54296875, "learning_rate": 0.0002365083583871534, "loss": 4.2287, "mean_token_accuracy": 0.29072198569774627, "num_tokens": 117431022.0, "step": 7225 }, { "entropy": 4.625004005432129, "epoch": 5.912065439672801, "grad_norm": 0.546875, "learning_rate": 0.00023619804060632754, "loss": 4.2607, "mean_token_accuracy": 0.28784058094024656, "num_tokens": 117512292.0, "step": 7230 }, { "entropy": 4.662065505981445, "epoch": 5.916155419222904, "grad_norm": 0.53125, "learning_rate": 0.00023588779887697399, "loss": 4.3257, "mean_token_accuracy": 0.28356009125709536, "num_tokens": 117593319.0, "step": 7235 }, { "entropy": 4.656345844268799, "epoch": 5.920245398773006, "grad_norm": 0.54296875, "learning_rate": 0.00023557763380716368, "loss": 4.2742, "mean_token_accuracy": 0.2881937503814697, "num_tokens": 117674915.0, "step": 7240 }, { "entropy": 4.6329052448272705, "epoch": 5.924335378323108, "grad_norm": 0.55078125, "learning_rate": 0.00023526754600481698, "loss": 4.2567, "mean_token_accuracy": 0.28872120976448057, "num_tokens": 117755423.0, "step": 7245 }, { "entropy": 4.633820486068726, "epoch": 5.928425357873211, "grad_norm": 0.5390625, "learning_rate": 0.00023495753607770322, "loss": 4.2653, "mean_token_accuracy": 0.28968707025051116, "num_tokens": 117837063.0, "step": 7250 }, { "entropy": 4.665157175064087, "epoch": 5.932515337423313, "grad_norm": 0.54296875, "learning_rate": 0.00023464760463343898, "loss": 4.2858, "mean_token_accuracy": 0.2853193461894989, "num_tokens": 117918645.0, "step": 7255 }, { "entropy": 4.674655628204346, "epoch": 5.936605316973415, "grad_norm": 0.52734375, "learning_rate": 0.0002343377522794867, "loss": 4.2913, "mean_token_accuracy": 0.2854655385017395, "num_tokens": 118000251.0, "step": 7260 }, { "entropy": 4.668893194198608, "epoch": 5.940695296523518, "grad_norm": 0.54296875, "learning_rate": 0.0002340279796231543, "loss": 4.3092, "mean_token_accuracy": 0.2822944611310959, "num_tokens": 118081885.0, "step": 7265 }, { "entropy": 4.641194486618042, "epoch": 5.9447852760736195, "grad_norm": 0.54296875, "learning_rate": 0.00023371828727159318, "loss": 4.2807, "mean_token_accuracy": 0.2859661728143692, "num_tokens": 118163353.0, "step": 7270 }, { "entropy": 4.635782861709595, "epoch": 5.948875255623722, "grad_norm": 0.5390625, "learning_rate": 0.00023340867583179726, "loss": 4.2667, "mean_token_accuracy": 0.2867343336343765, "num_tokens": 118244340.0, "step": 7275 }, { "entropy": 4.651674699783325, "epoch": 5.952965235173824, "grad_norm": 0.55859375, "learning_rate": 0.00023309914591060222, "loss": 4.2569, "mean_token_accuracy": 0.28722510039806365, "num_tokens": 118325693.0, "step": 7280 }, { "entropy": 4.662616443634033, "epoch": 5.957055214723926, "grad_norm": 0.53125, "learning_rate": 0.00023278969811468364, "loss": 4.3008, "mean_token_accuracy": 0.2840869665145874, "num_tokens": 118407389.0, "step": 7285 }, { "entropy": 4.653537845611572, "epoch": 5.961145194274029, "grad_norm": 0.51953125, "learning_rate": 0.00023248033305055631, "loss": 4.2656, "mean_token_accuracy": 0.288498067855835, "num_tokens": 118488444.0, "step": 7290 }, { "entropy": 4.6680567264556885, "epoch": 5.965235173824131, "grad_norm": 0.53515625, "learning_rate": 0.00023217105132457294, "loss": 4.2891, "mean_token_accuracy": 0.28600860238075254, "num_tokens": 118569019.0, "step": 7295 }, { "entropy": 4.649335289001465, "epoch": 5.969325153374233, "grad_norm": 0.546875, "learning_rate": 0.00023186185354292255, "loss": 4.2922, "mean_token_accuracy": 0.283039328455925, "num_tokens": 118650612.0, "step": 7300 }, { "entropy": 4.620884895324707, "epoch": 5.973415132924336, "grad_norm": 0.546875, "learning_rate": 0.00023155274031162992, "loss": 4.2433, "mean_token_accuracy": 0.28608475923538207, "num_tokens": 118732204.0, "step": 7305 }, { "entropy": 4.631065845489502, "epoch": 5.977505112474438, "grad_norm": 0.55078125, "learning_rate": 0.00023124371223655428, "loss": 4.2424, "mean_token_accuracy": 0.2899708330631256, "num_tokens": 118813521.0, "step": 7310 }, { "entropy": 4.6601556777954105, "epoch": 5.9815950920245395, "grad_norm": 0.54296875, "learning_rate": 0.00023093476992338746, "loss": 4.3219, "mean_token_accuracy": 0.28676224052906035, "num_tokens": 118894214.0, "step": 7315 }, { "entropy": 4.629721832275391, "epoch": 5.985685071574642, "grad_norm": 0.55078125, "learning_rate": 0.00023062591397765352, "loss": 4.2683, "mean_token_accuracy": 0.286335825920105, "num_tokens": 118974767.0, "step": 7320 }, { "entropy": 4.646107959747314, "epoch": 5.989775051124744, "grad_norm": 0.53125, "learning_rate": 0.00023031714500470736, "loss": 4.2582, "mean_token_accuracy": 0.289204353094101, "num_tokens": 119056421.0, "step": 7325 }, { "entropy": 4.67892165184021, "epoch": 5.993865030674847, "grad_norm": 0.54296875, "learning_rate": 0.00023000846360973293, "loss": 4.3071, "mean_token_accuracy": 0.28621521294116975, "num_tokens": 119137895.0, "step": 7330 }, { "entropy": 4.62217607498169, "epoch": 5.997955010224949, "grad_norm": 0.53515625, "learning_rate": 0.00022969987039774304, "loss": 4.2351, "mean_token_accuracy": 0.29026830196380615, "num_tokens": 119219324.0, "step": 7335 }, { "entropy": 4.557924588521321, "epoch": 6.001635991820041, "grad_norm": 0.5390625, "learning_rate": 0.00022939136597357752, "loss": 4.1003, "mean_token_accuracy": 0.2988593677679698, "num_tokens": 119285734.0, "step": 7340 }, { "entropy": 4.640133428573608, "epoch": 6.005725971370143, "grad_norm": 0.5546875, "learning_rate": 0.0002290829509419019, "loss": 4.1689, "mean_token_accuracy": 0.29808702766895295, "num_tokens": 119367334.0, "step": 7345 }, { "entropy": 4.6877274990081785, "epoch": 6.009815950920245, "grad_norm": 0.55859375, "learning_rate": 0.00022877462590720682, "loss": 4.1856, "mean_token_accuracy": 0.29781022369861604, "num_tokens": 119448560.0, "step": 7350 }, { "entropy": 4.645537090301514, "epoch": 6.013905930470347, "grad_norm": 0.5390625, "learning_rate": 0.0002284663914738065, "loss": 4.1626, "mean_token_accuracy": 0.29796794056892395, "num_tokens": 119530266.0, "step": 7355 }, { "entropy": 4.648176002502441, "epoch": 6.01799591002045, "grad_norm": 0.54296875, "learning_rate": 0.00022815824824583736, "loss": 4.2062, "mean_token_accuracy": 0.2947148323059082, "num_tokens": 119611209.0, "step": 7360 }, { "entropy": 4.660737085342407, "epoch": 6.022085889570552, "grad_norm": 0.55859375, "learning_rate": 0.00022785019682725738, "loss": 4.1772, "mean_token_accuracy": 0.29908466041088105, "num_tokens": 119692394.0, "step": 7365 }, { "entropy": 4.620302534103393, "epoch": 6.026175869120655, "grad_norm": 0.55859375, "learning_rate": 0.0002275422378218443, "loss": 4.141, "mean_token_accuracy": 0.2986477553844452, "num_tokens": 119772918.0, "step": 7370 }, { "entropy": 4.63349461555481, "epoch": 6.030265848670757, "grad_norm": 0.55859375, "learning_rate": 0.00022723437183319484, "loss": 4.1556, "mean_token_accuracy": 0.297368711233139, "num_tokens": 119854075.0, "step": 7375 }, { "entropy": 4.599965763092041, "epoch": 6.034355828220859, "grad_norm": 0.56640625, "learning_rate": 0.00022692659946472348, "loss": 4.1357, "mean_token_accuracy": 0.2992446571588516, "num_tokens": 119935680.0, "step": 7380 }, { "entropy": 4.68414535522461, "epoch": 6.0384458077709615, "grad_norm": 0.57421875, "learning_rate": 0.00022661892131966104, "loss": 4.2003, "mean_token_accuracy": 0.29441870748996735, "num_tokens": 120016121.0, "step": 7385 }, { "entropy": 4.615434455871582, "epoch": 6.0425357873210634, "grad_norm": 0.55078125, "learning_rate": 0.00022631133800105386, "loss": 4.1365, "mean_token_accuracy": 0.29822573661804197, "num_tokens": 120097380.0, "step": 7390 }, { "entropy": 4.618684959411621, "epoch": 6.046625766871165, "grad_norm": 0.5546875, "learning_rate": 0.00022600385011176238, "loss": 4.1919, "mean_token_accuracy": 0.29330260753631593, "num_tokens": 120178930.0, "step": 7395 }, { "entropy": 4.641580057144165, "epoch": 6.050715746421268, "grad_norm": 0.55078125, "learning_rate": 0.00022569645825445977, "loss": 4.154, "mean_token_accuracy": 0.2992739349603653, "num_tokens": 120259893.0, "step": 7400 }, { "entropy": 4.656009912490845, "epoch": 6.05480572597137, "grad_norm": 0.57421875, "learning_rate": 0.0002253891630316312, "loss": 4.1837, "mean_token_accuracy": 0.3003429532051086, "num_tokens": 120341060.0, "step": 7405 }, { "entropy": 4.6372456550598145, "epoch": 6.058895705521472, "grad_norm": 0.5625, "learning_rate": 0.0002250819650455726, "loss": 4.1852, "mean_token_accuracy": 0.2957141548395157, "num_tokens": 120422497.0, "step": 7410 }, { "entropy": 4.6140388488769535, "epoch": 6.062985685071575, "grad_norm": 0.58203125, "learning_rate": 0.0002247748648983889, "loss": 4.1417, "mean_token_accuracy": 0.3004630714654922, "num_tokens": 120503802.0, "step": 7415 }, { "entropy": 4.645229530334473, "epoch": 6.067075664621677, "grad_norm": 0.57421875, "learning_rate": 0.00022446786319199342, "loss": 4.1851, "mean_token_accuracy": 0.29406843781471254, "num_tokens": 120585382.0, "step": 7420 }, { "entropy": 4.670275115966797, "epoch": 6.071165644171779, "grad_norm": 0.5703125, "learning_rate": 0.00022416096052810686, "loss": 4.1929, "mean_token_accuracy": 0.29369590282440183, "num_tokens": 120666693.0, "step": 7425 }, { "entropy": 4.606317806243896, "epoch": 6.0752556237218815, "grad_norm": 0.5546875, "learning_rate": 0.00022385415750825522, "loss": 4.1617, "mean_token_accuracy": 0.2951271295547485, "num_tokens": 120748174.0, "step": 7430 }, { "entropy": 4.684191131591797, "epoch": 6.0793456032719835, "grad_norm": 0.5703125, "learning_rate": 0.0002235474547337697, "loss": 4.2403, "mean_token_accuracy": 0.29235540330410004, "num_tokens": 120829816.0, "step": 7435 }, { "entropy": 4.664311695098877, "epoch": 6.083435582822086, "grad_norm": 0.56640625, "learning_rate": 0.00022324085280578482, "loss": 4.2127, "mean_token_accuracy": 0.29453728795051576, "num_tokens": 120911135.0, "step": 7440 }, { "entropy": 4.60966157913208, "epoch": 6.087525562372188, "grad_norm": 0.546875, "learning_rate": 0.0002229343523252374, "loss": 4.1495, "mean_token_accuracy": 0.3004549354314804, "num_tokens": 120992822.0, "step": 7445 }, { "entropy": 4.592287302017212, "epoch": 6.09161554192229, "grad_norm": 0.57421875, "learning_rate": 0.00022262795389286545, "loss": 4.1287, "mean_token_accuracy": 0.2980617344379425, "num_tokens": 121074173.0, "step": 7450 }, { "entropy": 4.665419006347657, "epoch": 6.095705521472393, "grad_norm": 0.578125, "learning_rate": 0.00022232165810920712, "loss": 4.1985, "mean_token_accuracy": 0.2935770660638809, "num_tokens": 121155315.0, "step": 7455 }, { "entropy": 4.608814811706543, "epoch": 6.099795501022495, "grad_norm": 0.55078125, "learning_rate": 0.00022201546557459905, "loss": 4.13, "mean_token_accuracy": 0.29839873909950254, "num_tokens": 121236194.0, "step": 7460 }, { "entropy": 4.613042497634888, "epoch": 6.103885480572597, "grad_norm": 0.56640625, "learning_rate": 0.0002217093768891758, "loss": 4.1416, "mean_token_accuracy": 0.29823012351989747, "num_tokens": 121317472.0, "step": 7465 }, { "entropy": 4.622537612915039, "epoch": 6.1079754601227, "grad_norm": 0.5625, "learning_rate": 0.0002214033926528684, "loss": 4.1581, "mean_token_accuracy": 0.297630313038826, "num_tokens": 121398665.0, "step": 7470 }, { "entropy": 4.582029914855957, "epoch": 6.112065439672802, "grad_norm": 0.55859375, "learning_rate": 0.00022109751346540286, "loss": 4.1468, "mean_token_accuracy": 0.299282631278038, "num_tokens": 121480323.0, "step": 7475 }, { "entropy": 4.640454435348511, "epoch": 6.1161554192229035, "grad_norm": 0.55859375, "learning_rate": 0.00022079173992629964, "loss": 4.2032, "mean_token_accuracy": 0.2916122227907181, "num_tokens": 121561933.0, "step": 7480 }, { "entropy": 4.657577037811279, "epoch": 6.120245398773006, "grad_norm": 0.57421875, "learning_rate": 0.0002204860726348718, "loss": 4.1737, "mean_token_accuracy": 0.2955718725919724, "num_tokens": 121642831.0, "step": 7485 }, { "entropy": 4.592429733276367, "epoch": 6.124335378323108, "grad_norm": 0.546875, "learning_rate": 0.00022018051219022445, "loss": 4.1309, "mean_token_accuracy": 0.29900096356868744, "num_tokens": 121724112.0, "step": 7490 }, { "entropy": 4.696337699890137, "epoch": 6.12842535787321, "grad_norm": 0.58203125, "learning_rate": 0.00021987505919125324, "loss": 4.2559, "mean_token_accuracy": 0.28790113627910613, "num_tokens": 121805619.0, "step": 7495 }, { "entropy": 4.643082523345948, "epoch": 6.132515337423313, "grad_norm": 0.5703125, "learning_rate": 0.00021956971423664297, "loss": 4.1577, "mean_token_accuracy": 0.29617200791835785, "num_tokens": 121886839.0, "step": 7500 }, { "epoch": 6.132515337423313, "eval_entropy": 4.493859815781641, "eval_loss": 4.602505683898926, "eval_mean_token_accuracy": 0.26807936942600374, "eval_num_tokens": 121886839.0, "eval_runtime": 68.997, "eval_samples_per_second": 59.959, "eval_steps_per_second": 7.508, "step": 7500 }, { "entropy": 4.648884105682373, "epoch": 6.136605316973415, "grad_norm": 0.55078125, "learning_rate": 0.0002192644779248669, "loss": 4.1933, "mean_token_accuracy": 0.2966649353504181, "num_tokens": 121968440.0, "step": 7505 }, { "entropy": 4.658957481384277, "epoch": 6.140695296523518, "grad_norm": 0.5703125, "learning_rate": 0.0002189593508541855, "loss": 4.2094, "mean_token_accuracy": 0.291450560092926, "num_tokens": 122049379.0, "step": 7510 }, { "entropy": 4.6170977592468265, "epoch": 6.14478527607362, "grad_norm": 0.55859375, "learning_rate": 0.00021865433362264469, "loss": 4.1624, "mean_token_accuracy": 0.2949069231748581, "num_tokens": 122131019.0, "step": 7515 }, { "entropy": 4.6723487854003904, "epoch": 6.148875255623722, "grad_norm": 0.57421875, "learning_rate": 0.0002183494268280754, "loss": 4.2045, "mean_token_accuracy": 0.29481164216995237, "num_tokens": 122211567.0, "step": 7520 }, { "entropy": 4.648819065093994, "epoch": 6.152965235173824, "grad_norm": 0.5625, "learning_rate": 0.00021804463106809218, "loss": 4.1938, "mean_token_accuracy": 0.2934700667858124, "num_tokens": 122293275.0, "step": 7525 }, { "entropy": 4.677602100372314, "epoch": 6.157055214723926, "grad_norm": 0.5703125, "learning_rate": 0.00021773994694009174, "loss": 4.2429, "mean_token_accuracy": 0.2915168523788452, "num_tokens": 122374710.0, "step": 7530 }, { "entropy": 4.623123359680176, "epoch": 6.161145194274028, "grad_norm": 0.58203125, "learning_rate": 0.000217435375041252, "loss": 4.1725, "mean_token_accuracy": 0.29263406097888944, "num_tokens": 122456312.0, "step": 7535 }, { "entropy": 4.691426420211792, "epoch": 6.165235173824131, "grad_norm": 0.57421875, "learning_rate": 0.0002171309159685312, "loss": 4.2351, "mean_token_accuracy": 0.29385215044021606, "num_tokens": 122537639.0, "step": 7540 }, { "entropy": 4.59920711517334, "epoch": 6.169325153374233, "grad_norm": 0.5625, "learning_rate": 0.00021682657031866598, "loss": 4.1465, "mean_token_accuracy": 0.29982012808322905, "num_tokens": 122618977.0, "step": 7545 }, { "entropy": 4.664438199996948, "epoch": 6.173415132924335, "grad_norm": 0.58203125, "learning_rate": 0.00021652233868817107, "loss": 4.2009, "mean_token_accuracy": 0.2920301049947739, "num_tokens": 122700197.0, "step": 7550 }, { "entropy": 4.637759447097778, "epoch": 6.177505112474438, "grad_norm": 0.57421875, "learning_rate": 0.0002162182216733377, "loss": 4.1855, "mean_token_accuracy": 0.2954164147377014, "num_tokens": 122781787.0, "step": 7555 }, { "entropy": 4.621495723724365, "epoch": 6.18159509202454, "grad_norm": 0.5859375, "learning_rate": 0.00021591421987023212, "loss": 4.1887, "mean_token_accuracy": 0.2953436881303787, "num_tokens": 122863409.0, "step": 7560 }, { "entropy": 4.6568629264831545, "epoch": 6.1856850715746425, "grad_norm": 0.56640625, "learning_rate": 0.00021561033387469504, "loss": 4.1919, "mean_token_accuracy": 0.29381059408187865, "num_tokens": 122944893.0, "step": 7565 }, { "entropy": 4.6462596416473385, "epoch": 6.1897750511247445, "grad_norm": 0.5703125, "learning_rate": 0.0002153065642823402, "loss": 4.1909, "mean_token_accuracy": 0.29446708858013154, "num_tokens": 123026529.0, "step": 7570 }, { "entropy": 4.622148132324218, "epoch": 6.193865030674846, "grad_norm": 0.55859375, "learning_rate": 0.00021500291168855302, "loss": 4.1599, "mean_token_accuracy": 0.29755058586597444, "num_tokens": 123107623.0, "step": 7575 }, { "entropy": 4.611489152908325, "epoch": 6.197955010224949, "grad_norm": 0.57421875, "learning_rate": 0.00021469937668848983, "loss": 4.1714, "mean_token_accuracy": 0.2956318467855453, "num_tokens": 123189033.0, "step": 7580 }, { "entropy": 4.619315338134766, "epoch": 6.202044989775051, "grad_norm": 0.55859375, "learning_rate": 0.00021439595987707616, "loss": 4.1718, "mean_token_accuracy": 0.29601856768131257, "num_tokens": 123270612.0, "step": 7585 }, { "entropy": 4.61763186454773, "epoch": 6.206134969325153, "grad_norm": 0.5625, "learning_rate": 0.00021409266184900622, "loss": 4.1531, "mean_token_accuracy": 0.2972570687532425, "num_tokens": 123352246.0, "step": 7590 }, { "entropy": 4.6118762493133545, "epoch": 6.210224948875256, "grad_norm": 0.5703125, "learning_rate": 0.0002137894831987413, "loss": 4.1641, "mean_token_accuracy": 0.3001962542533875, "num_tokens": 123432957.0, "step": 7595 }, { "entropy": 4.649173307418823, "epoch": 6.214314928425358, "grad_norm": 0.5625, "learning_rate": 0.0002134864245205086, "loss": 4.192, "mean_token_accuracy": 0.2924411058425903, "num_tokens": 123514646.0, "step": 7600 }, { "entropy": 4.604610347747803, "epoch": 6.21840490797546, "grad_norm": 0.53515625, "learning_rate": 0.00021318348640830026, "loss": 4.1563, "mean_token_accuracy": 0.2959522545337677, "num_tokens": 123596177.0, "step": 7605 }, { "entropy": 4.603279733657837, "epoch": 6.222494887525563, "grad_norm": 0.578125, "learning_rate": 0.00021288066945587215, "loss": 4.1878, "mean_token_accuracy": 0.2935083985328674, "num_tokens": 123677119.0, "step": 7610 }, { "entropy": 4.61686601638794, "epoch": 6.2265848670756645, "grad_norm": 0.5625, "learning_rate": 0.0002125779742567426, "loss": 4.1567, "mean_token_accuracy": 0.29632003605365753, "num_tokens": 123758736.0, "step": 7615 }, { "entropy": 4.627933311462402, "epoch": 6.230674846625767, "grad_norm": 0.578125, "learning_rate": 0.00021227540140419127, "loss": 4.1864, "mean_token_accuracy": 0.2957927197217941, "num_tokens": 123839650.0, "step": 7620 }, { "entropy": 4.627634429931641, "epoch": 6.234764826175869, "grad_norm": 0.578125, "learning_rate": 0.00021197295149125823, "loss": 4.1583, "mean_token_accuracy": 0.2934717059135437, "num_tokens": 123920860.0, "step": 7625 }, { "entropy": 4.581482791900635, "epoch": 6.238854805725971, "grad_norm": 0.5625, "learning_rate": 0.0002116706251107422, "loss": 4.1388, "mean_token_accuracy": 0.29914712607860566, "num_tokens": 124002495.0, "step": 7630 }, { "entropy": 4.644417238235474, "epoch": 6.242944785276074, "grad_norm": 0.5625, "learning_rate": 0.00021136842285520018, "loss": 4.1849, "mean_token_accuracy": 0.2904126465320587, "num_tokens": 124083701.0, "step": 7635 }, { "entropy": 4.627776765823365, "epoch": 6.247034764826176, "grad_norm": 0.56640625, "learning_rate": 0.00021106634531694584, "loss": 4.1811, "mean_token_accuracy": 0.2970398277044296, "num_tokens": 124164634.0, "step": 7640 }, { "entropy": 4.636410713195801, "epoch": 6.251124744376278, "grad_norm": 0.57421875, "learning_rate": 0.00021076439308804806, "loss": 4.1914, "mean_token_accuracy": 0.2951155036687851, "num_tokens": 124245810.0, "step": 7645 }, { "entropy": 4.628436613082886, "epoch": 6.255214723926381, "grad_norm": 0.5703125, "learning_rate": 0.00021046256676033044, "loss": 4.1678, "mean_token_accuracy": 0.2924444407224655, "num_tokens": 124326204.0, "step": 7650 }, { "entropy": 4.603310537338257, "epoch": 6.259304703476483, "grad_norm": 0.5546875, "learning_rate": 0.0002101608669253698, "loss": 4.1537, "mean_token_accuracy": 0.29234030842781067, "num_tokens": 124407551.0, "step": 7655 }, { "entropy": 4.647319030761719, "epoch": 6.2633946830265845, "grad_norm": 0.56640625, "learning_rate": 0.00020985929417449478, "loss": 4.208, "mean_token_accuracy": 0.2910054087638855, "num_tokens": 124489282.0, "step": 7660 }, { "entropy": 4.637158536911011, "epoch": 6.267484662576687, "grad_norm": 0.56640625, "learning_rate": 0.00020955784909878523, "loss": 4.1895, "mean_token_accuracy": 0.29610815942287444, "num_tokens": 124570935.0, "step": 7665 }, { "entropy": 4.59713306427002, "epoch": 6.271574642126789, "grad_norm": 0.58203125, "learning_rate": 0.00020925653228907065, "loss": 4.1241, "mean_token_accuracy": 0.30358612835407256, "num_tokens": 124652264.0, "step": 7670 }, { "entropy": 4.637485027313232, "epoch": 6.275664621676891, "grad_norm": 0.58203125, "learning_rate": 0.000208955344335929, "loss": 4.1855, "mean_token_accuracy": 0.2927857875823975, "num_tokens": 124733562.0, "step": 7675 }, { "entropy": 4.656131362915039, "epoch": 6.279754601226994, "grad_norm": 0.609375, "learning_rate": 0.00020865428582968594, "loss": 4.2059, "mean_token_accuracy": 0.2938678652048111, "num_tokens": 124815167.0, "step": 7680 }, { "entropy": 4.6411501407623295, "epoch": 6.283844580777096, "grad_norm": 0.56640625, "learning_rate": 0.00020835335736041333, "loss": 4.1906, "mean_token_accuracy": 0.29545426964759824, "num_tokens": 124896290.0, "step": 7685 }, { "entropy": 4.665257549285888, "epoch": 6.287934560327199, "grad_norm": 0.5703125, "learning_rate": 0.00020805255951792794, "loss": 4.1981, "mean_token_accuracy": 0.294184473156929, "num_tokens": 124977851.0, "step": 7690 }, { "entropy": 4.595583057403564, "epoch": 6.292024539877301, "grad_norm": 0.55859375, "learning_rate": 0.00020775189289179086, "loss": 4.161, "mean_token_accuracy": 0.2971080541610718, "num_tokens": 125059040.0, "step": 7695 }, { "entropy": 4.609917163848877, "epoch": 6.296114519427403, "grad_norm": 0.56640625, "learning_rate": 0.00020745135807130567, "loss": 4.1799, "mean_token_accuracy": 0.2931637465953827, "num_tokens": 125140711.0, "step": 7700 }, { "entropy": 4.604016923904419, "epoch": 6.3002044989775055, "grad_norm": 0.55859375, "learning_rate": 0.0002071509556455179, "loss": 4.1468, "mean_token_accuracy": 0.29632535576820374, "num_tokens": 125222040.0, "step": 7705 }, { "entropy": 4.639315271377564, "epoch": 6.304294478527607, "grad_norm": 0.5625, "learning_rate": 0.00020685068620321335, "loss": 4.1939, "mean_token_accuracy": 0.2926967263221741, "num_tokens": 125303695.0, "step": 7710 }, { "entropy": 4.626247310638428, "epoch": 6.308384458077709, "grad_norm": 0.5703125, "learning_rate": 0.00020655055033291737, "loss": 4.1801, "mean_token_accuracy": 0.29579761028289797, "num_tokens": 125384866.0, "step": 7715 }, { "entropy": 4.639872026443482, "epoch": 6.312474437627812, "grad_norm": 0.56640625, "learning_rate": 0.00020625054862289326, "loss": 4.1757, "mean_token_accuracy": 0.2947367250919342, "num_tokens": 125465737.0, "step": 7720 }, { "entropy": 4.622911167144776, "epoch": 6.316564417177914, "grad_norm": 0.57421875, "learning_rate": 0.00020595068166114178, "loss": 4.1679, "mean_token_accuracy": 0.2928102195262909, "num_tokens": 125546796.0, "step": 7725 }, { "entropy": 4.600707101821899, "epoch": 6.320654396728016, "grad_norm": 0.55859375, "learning_rate": 0.00020565095003539913, "loss": 4.1583, "mean_token_accuracy": 0.29655219316482545, "num_tokens": 125628400.0, "step": 7730 }, { "entropy": 4.633536911010742, "epoch": 6.324744376278119, "grad_norm": 0.56640625, "learning_rate": 0.00020535135433313645, "loss": 4.2119, "mean_token_accuracy": 0.29211550354957583, "num_tokens": 125709691.0, "step": 7735 }, { "entropy": 4.683541870117187, "epoch": 6.328834355828221, "grad_norm": 0.5703125, "learning_rate": 0.0002050518951415587, "loss": 4.2172, "mean_token_accuracy": 0.291774383187294, "num_tokens": 125791358.0, "step": 7740 }, { "entropy": 4.6176745891571045, "epoch": 6.332924335378323, "grad_norm": 0.578125, "learning_rate": 0.00020475257304760274, "loss": 4.1675, "mean_token_accuracy": 0.2950634151697159, "num_tokens": 125872646.0, "step": 7745 }, { "entropy": 4.633988571166992, "epoch": 6.3370143149284255, "grad_norm": 0.58203125, "learning_rate": 0.00020445338863793726, "loss": 4.165, "mean_token_accuracy": 0.29662005603313446, "num_tokens": 125954009.0, "step": 7750 }, { "entropy": 4.658006715774536, "epoch": 6.341104294478527, "grad_norm": 0.58984375, "learning_rate": 0.00020415434249896076, "loss": 4.2228, "mean_token_accuracy": 0.2909741550683975, "num_tokens": 126035065.0, "step": 7755 }, { "entropy": 4.649401140213013, "epoch": 6.34519427402863, "grad_norm": 0.578125, "learning_rate": 0.00020385543521680088, "loss": 4.212, "mean_token_accuracy": 0.2928553432226181, "num_tokens": 126115883.0, "step": 7760 }, { "entropy": 4.629928159713745, "epoch": 6.349284253578732, "grad_norm": 0.5703125, "learning_rate": 0.000203556667377313, "loss": 4.1552, "mean_token_accuracy": 0.2983356773853302, "num_tokens": 126197454.0, "step": 7765 }, { "entropy": 4.5797240257263185, "epoch": 6.353374233128834, "grad_norm": 0.57421875, "learning_rate": 0.00020325803956607941, "loss": 4.1483, "mean_token_accuracy": 0.2959419935941696, "num_tokens": 126277838.0, "step": 7770 }, { "entropy": 4.617911386489868, "epoch": 6.357464212678937, "grad_norm": 0.56640625, "learning_rate": 0.00020295955236840753, "loss": 4.1823, "mean_token_accuracy": 0.2939877390861511, "num_tokens": 126359455.0, "step": 7775 }, { "entropy": 4.615458869934082, "epoch": 6.361554192229039, "grad_norm": 0.5546875, "learning_rate": 0.00020266120636932961, "loss": 4.1592, "mean_token_accuracy": 0.2978530526161194, "num_tokens": 126440743.0, "step": 7780 }, { "entropy": 4.623746299743653, "epoch": 6.365644171779141, "grad_norm": 0.57421875, "learning_rate": 0.00020236300215360103, "loss": 4.1856, "mean_token_accuracy": 0.295445054769516, "num_tokens": 126521551.0, "step": 7785 }, { "entropy": 4.632915449142456, "epoch": 6.369734151329244, "grad_norm": 0.56640625, "learning_rate": 0.00020206494030569906, "loss": 4.209, "mean_token_accuracy": 0.29359435439109804, "num_tokens": 126602538.0, "step": 7790 }, { "entropy": 4.617979192733765, "epoch": 6.3738241308793455, "grad_norm": 0.5703125, "learning_rate": 0.0002017670214098223, "loss": 4.1817, "mean_token_accuracy": 0.2976475328207016, "num_tokens": 126684129.0, "step": 7795 }, { "entropy": 4.6348193168640135, "epoch": 6.3779141104294474, "grad_norm": 0.59375, "learning_rate": 0.0002014692460498887, "loss": 4.184, "mean_token_accuracy": 0.29544466733932495, "num_tokens": 126765425.0, "step": 7800 }, { "entropy": 4.597709035873413, "epoch": 6.38200408997955, "grad_norm": 0.5703125, "learning_rate": 0.00020117161480953527, "loss": 4.133, "mean_token_accuracy": 0.30076943039894105, "num_tokens": 126846974.0, "step": 7805 }, { "entropy": 4.638014554977417, "epoch": 6.386094069529652, "grad_norm": 0.57421875, "learning_rate": 0.00020087412827211648, "loss": 4.2053, "mean_token_accuracy": 0.29396226406097414, "num_tokens": 126927816.0, "step": 7810 }, { "entropy": 4.651465606689453, "epoch": 6.390184049079755, "grad_norm": 0.578125, "learning_rate": 0.00020057678702070303, "loss": 4.2119, "mean_token_accuracy": 0.29179295599460603, "num_tokens": 127009558.0, "step": 7815 }, { "entropy": 4.601522493362427, "epoch": 6.394274028629857, "grad_norm": 0.56640625, "learning_rate": 0.00020027959163808096, "loss": 4.1539, "mean_token_accuracy": 0.2957317978143692, "num_tokens": 127091057.0, "step": 7820 }, { "entropy": 4.635461473464966, "epoch": 6.398364008179959, "grad_norm": 0.5625, "learning_rate": 0.00019998254270675055, "loss": 4.1865, "mean_token_accuracy": 0.2949948489665985, "num_tokens": 127172715.0, "step": 7825 }, { "entropy": 4.598548030853271, "epoch": 6.402453987730062, "grad_norm": 0.5625, "learning_rate": 0.0001996856408089246, "loss": 4.1467, "mean_token_accuracy": 0.29674625396728516, "num_tokens": 127254110.0, "step": 7830 }, { "entropy": 4.630077743530274, "epoch": 6.406543967280164, "grad_norm": 0.57421875, "learning_rate": 0.0001993888865265282, "loss": 4.2153, "mean_token_accuracy": 0.2944152057170868, "num_tokens": 127335623.0, "step": 7835 }, { "entropy": 4.659919881820679, "epoch": 6.4106339468302656, "grad_norm": 0.56640625, "learning_rate": 0.00019909228044119704, "loss": 4.1838, "mean_token_accuracy": 0.2958721786737442, "num_tokens": 127416638.0, "step": 7840 }, { "entropy": 4.613992166519165, "epoch": 6.414723926380368, "grad_norm": 0.58203125, "learning_rate": 0.00019879582313427597, "loss": 4.1796, "mean_token_accuracy": 0.2975289076566696, "num_tokens": 127498041.0, "step": 7845 }, { "entropy": 4.617488288879395, "epoch": 6.41881390593047, "grad_norm": 0.56640625, "learning_rate": 0.00019849951518681865, "loss": 4.1768, "mean_token_accuracy": 0.2950626343488693, "num_tokens": 127579688.0, "step": 7850 }, { "entropy": 4.645755434036255, "epoch": 6.422903885480572, "grad_norm": 0.57421875, "learning_rate": 0.00019820335717958586, "loss": 4.181, "mean_token_accuracy": 0.2957373410463333, "num_tokens": 127660848.0, "step": 7855 }, { "entropy": 4.635202312469483, "epoch": 6.426993865030675, "grad_norm": 0.5859375, "learning_rate": 0.00019790734969304442, "loss": 4.2056, "mean_token_accuracy": 0.295372998714447, "num_tokens": 127741922.0, "step": 7860 }, { "entropy": 4.619224309921265, "epoch": 6.431083844580777, "grad_norm": 0.59375, "learning_rate": 0.00019761149330736616, "loss": 4.1515, "mean_token_accuracy": 0.3010107815265656, "num_tokens": 127823464.0, "step": 7865 }, { "entropy": 4.6489561080932615, "epoch": 6.43517382413088, "grad_norm": 0.578125, "learning_rate": 0.000197315788602427, "loss": 4.2052, "mean_token_accuracy": 0.2913580656051636, "num_tokens": 127905096.0, "step": 7870 }, { "entropy": 4.618567276000976, "epoch": 6.439263803680982, "grad_norm": 0.55078125, "learning_rate": 0.00019702023615780507, "loss": 4.173, "mean_token_accuracy": 0.29256437718868256, "num_tokens": 127986453.0, "step": 7875 }, { "entropy": 4.613063192367553, "epoch": 6.443353783231084, "grad_norm": 0.5703125, "learning_rate": 0.0001967248365527805, "loss": 4.1691, "mean_token_accuracy": 0.29769307076931, "num_tokens": 128068132.0, "step": 7880 }, { "entropy": 4.622210311889648, "epoch": 6.4474437627811865, "grad_norm": 0.5859375, "learning_rate": 0.00019642959036633378, "loss": 4.2138, "mean_token_accuracy": 0.2922559976577759, "num_tokens": 128148951.0, "step": 7885 }, { "entropy": 4.670883893966675, "epoch": 6.451533742331288, "grad_norm": 0.5703125, "learning_rate": 0.00019613449817714452, "loss": 4.2302, "mean_token_accuracy": 0.29033604860305784, "num_tokens": 128230323.0, "step": 7890 }, { "entropy": 4.634058284759521, "epoch": 6.45562372188139, "grad_norm": 0.5625, "learning_rate": 0.00019583956056359075, "loss": 4.1551, "mean_token_accuracy": 0.29593655467033386, "num_tokens": 128310753.0, "step": 7895 }, { "entropy": 4.596018123626709, "epoch": 6.459713701431493, "grad_norm": 0.5859375, "learning_rate": 0.00019554477810374715, "loss": 4.1523, "mean_token_accuracy": 0.29893714785575864, "num_tokens": 128392364.0, "step": 7900 }, { "entropy": 4.612285423278808, "epoch": 6.463803680981595, "grad_norm": 0.56640625, "learning_rate": 0.00019525015137538478, "loss": 4.1757, "mean_token_accuracy": 0.29512579143047335, "num_tokens": 128473958.0, "step": 7905 }, { "entropy": 4.616873455047608, "epoch": 6.467893660531697, "grad_norm": 0.578125, "learning_rate": 0.00019495568095596915, "loss": 4.199, "mean_token_accuracy": 0.2921194851398468, "num_tokens": 128555575.0, "step": 7910 }, { "entropy": 4.621131753921508, "epoch": 6.4719836400818, "grad_norm": 0.55859375, "learning_rate": 0.0001946613674226595, "loss": 4.1723, "mean_token_accuracy": 0.2999284893274307, "num_tokens": 128636378.0, "step": 7915 }, { "entropy": 4.62133641242981, "epoch": 6.476073619631902, "grad_norm": 0.5703125, "learning_rate": 0.00019436721135230746, "loss": 4.2001, "mean_token_accuracy": 0.29320008456707003, "num_tokens": 128716808.0, "step": 7920 }, { "entropy": 4.60214638710022, "epoch": 6.480163599182004, "grad_norm": 0.56640625, "learning_rate": 0.00019407321332145632, "loss": 4.1756, "mean_token_accuracy": 0.2955467015504837, "num_tokens": 128798320.0, "step": 7925 }, { "entropy": 4.5964577198028564, "epoch": 6.4842535787321065, "grad_norm": 0.5625, "learning_rate": 0.00019377937390633926, "loss": 4.1667, "mean_token_accuracy": 0.29799906015396116, "num_tokens": 128879793.0, "step": 7930 }, { "entropy": 4.648832702636719, "epoch": 6.488343558282208, "grad_norm": 0.56640625, "learning_rate": 0.00019348569368287882, "loss": 4.2082, "mean_token_accuracy": 0.2915367305278778, "num_tokens": 128961365.0, "step": 7935 }, { "entropy": 4.634428548812866, "epoch": 6.492433537832311, "grad_norm": 0.56640625, "learning_rate": 0.00019319217322668554, "loss": 4.1729, "mean_token_accuracy": 0.29400024712085726, "num_tokens": 129042815.0, "step": 7940 }, { "entropy": 4.618898391723633, "epoch": 6.496523517382413, "grad_norm": 0.57421875, "learning_rate": 0.00019289881311305656, "loss": 4.1855, "mean_token_accuracy": 0.29244156181812286, "num_tokens": 129123898.0, "step": 7945 }, { "entropy": 4.641170310974121, "epoch": 6.500613496932515, "grad_norm": 0.578125, "learning_rate": 0.00019260561391697505, "loss": 4.1964, "mean_token_accuracy": 0.2957218289375305, "num_tokens": 129205383.0, "step": 7950 }, { "entropy": 4.649530363082886, "epoch": 6.504703476482618, "grad_norm": 0.58203125, "learning_rate": 0.0001923125762131086, "loss": 4.223, "mean_token_accuracy": 0.2916749507188797, "num_tokens": 129286990.0, "step": 7955 }, { "entropy": 4.629819250106811, "epoch": 6.50879345603272, "grad_norm": 0.57421875, "learning_rate": 0.00019201970057580836, "loss": 4.2021, "mean_token_accuracy": 0.2946648269891739, "num_tokens": 129368224.0, "step": 7960 }, { "entropy": 4.634172868728638, "epoch": 6.512883435582822, "grad_norm": 0.57421875, "learning_rate": 0.0001917269875791077, "loss": 4.1929, "mean_token_accuracy": 0.293327796459198, "num_tokens": 129449460.0, "step": 7965 }, { "entropy": 4.629204607009887, "epoch": 6.516973415132925, "grad_norm": 0.5703125, "learning_rate": 0.00019143443779672155, "loss": 4.174, "mean_token_accuracy": 0.29639307856559755, "num_tokens": 129531075.0, "step": 7970 }, { "entropy": 4.606035041809082, "epoch": 6.5210633946830265, "grad_norm": 0.55859375, "learning_rate": 0.00019114205180204436, "loss": 4.1707, "mean_token_accuracy": 0.29842317998409273, "num_tokens": 129612429.0, "step": 7975 }, { "entropy": 4.6082422733306885, "epoch": 6.5251533742331285, "grad_norm": 0.5625, "learning_rate": 0.00019084983016815017, "loss": 4.1542, "mean_token_accuracy": 0.296645450592041, "num_tokens": 129693880.0, "step": 7980 }, { "entropy": 4.63595724105835, "epoch": 6.529243353783231, "grad_norm": 0.58203125, "learning_rate": 0.00019055777346779057, "loss": 4.1999, "mean_token_accuracy": 0.29399033784866335, "num_tokens": 129775021.0, "step": 7985 }, { "entropy": 4.625579357147217, "epoch": 6.533333333333333, "grad_norm": 0.57421875, "learning_rate": 0.00019026588227339386, "loss": 4.1488, "mean_token_accuracy": 0.2981957882642746, "num_tokens": 129855728.0, "step": 7990 }, { "entropy": 4.625676536560059, "epoch": 6.537423312883435, "grad_norm": 0.55859375, "learning_rate": 0.00018997415715706412, "loss": 4.1914, "mean_token_accuracy": 0.29610956013202666, "num_tokens": 129937223.0, "step": 7995 }, { "entropy": 4.599562263488769, "epoch": 6.541513292433538, "grad_norm": 0.55078125, "learning_rate": 0.00018968259869057958, "loss": 4.1704, "mean_token_accuracy": 0.29832259118556975, "num_tokens": 130018817.0, "step": 8000 }, { "epoch": 6.541513292433538, "eval_entropy": 4.450945293581164, "eval_loss": 4.591839790344238, "eval_mean_token_accuracy": 0.26931912057878427, "eval_num_tokens": 130018817.0, "eval_runtime": 66.2215, "eval_samples_per_second": 62.472, "eval_steps_per_second": 7.822, "step": 8000 }, { "entropy": 4.5967058658599855, "epoch": 6.54560327198364, "grad_norm": 0.58984375, "learning_rate": 0.00018939120744539234, "loss": 4.1663, "mean_token_accuracy": 0.2959628880023956, "num_tokens": 130099693.0, "step": 8005 }, { "entropy": 4.6327619552612305, "epoch": 6.549693251533743, "grad_norm": 0.578125, "learning_rate": 0.00018909998399262634, "loss": 4.1892, "mean_token_accuracy": 0.29458103477954867, "num_tokens": 130181334.0, "step": 8010 }, { "entropy": 4.6078596115112305, "epoch": 6.553783231083845, "grad_norm": 0.55859375, "learning_rate": 0.00018880892890307677, "loss": 4.1742, "mean_token_accuracy": 0.2972289353609085, "num_tokens": 130262901.0, "step": 8015 }, { "entropy": 4.6618578910827635, "epoch": 6.557873210633947, "grad_norm": 0.5703125, "learning_rate": 0.00018851804274720886, "loss": 4.2359, "mean_token_accuracy": 0.285745832324028, "num_tokens": 130344416.0, "step": 8020 }, { "entropy": 4.594387817382812, "epoch": 6.561963190184049, "grad_norm": 0.578125, "learning_rate": 0.0001882273260951568, "loss": 4.1458, "mean_token_accuracy": 0.30107806622982025, "num_tokens": 130425744.0, "step": 8025 }, { "entropy": 4.619293642044068, "epoch": 6.566053169734151, "grad_norm": 0.57421875, "learning_rate": 0.00018793677951672234, "loss": 4.1804, "mean_token_accuracy": 0.2974590629339218, "num_tokens": 130507284.0, "step": 8030 }, { "entropy": 4.61993408203125, "epoch": 6.570143149284253, "grad_norm": 0.58203125, "learning_rate": 0.00018764640358137407, "loss": 4.1754, "mean_token_accuracy": 0.29270164370536805, "num_tokens": 130588515.0, "step": 8035 }, { "entropy": 4.614811897277832, "epoch": 6.574233128834356, "grad_norm": 0.58203125, "learning_rate": 0.00018735619885824618, "loss": 4.1818, "mean_token_accuracy": 0.29634890258312224, "num_tokens": 130670149.0, "step": 8040 }, { "entropy": 4.640796661376953, "epoch": 6.578323108384458, "grad_norm": 0.578125, "learning_rate": 0.00018706616591613695, "loss": 4.2146, "mean_token_accuracy": 0.2938166171312332, "num_tokens": 130751291.0, "step": 8045 }, { "entropy": 4.619365787506103, "epoch": 6.58241308793456, "grad_norm": 0.56640625, "learning_rate": 0.00018677630532350835, "loss": 4.1915, "mean_token_accuracy": 0.2921881675720215, "num_tokens": 130832581.0, "step": 8050 }, { "entropy": 4.6322063446044925, "epoch": 6.586503067484663, "grad_norm": 0.57421875, "learning_rate": 0.0001864866176484844, "loss": 4.2188, "mean_token_accuracy": 0.29202547669410706, "num_tokens": 130913453.0, "step": 8055 }, { "entropy": 4.6329405307769775, "epoch": 6.590593047034765, "grad_norm": 0.5625, "learning_rate": 0.00018619710345885016, "loss": 4.1739, "mean_token_accuracy": 0.2964888662099838, "num_tokens": 130995082.0, "step": 8060 }, { "entropy": 4.630719995498657, "epoch": 6.5946830265848675, "grad_norm": 0.5703125, "learning_rate": 0.0001859077633220506, "loss": 4.2038, "mean_token_accuracy": 0.292615732550621, "num_tokens": 131076630.0, "step": 8065 }, { "entropy": 4.619901275634765, "epoch": 6.598773006134969, "grad_norm": 0.58203125, "learning_rate": 0.00018561859780518987, "loss": 4.1899, "mean_token_accuracy": 0.2935108572244644, "num_tokens": 131158275.0, "step": 8070 }, { "entropy": 4.6328568935394285, "epoch": 6.602862985685071, "grad_norm": 0.57421875, "learning_rate": 0.00018532960747502937, "loss": 4.1928, "mean_token_accuracy": 0.29357501268386843, "num_tokens": 131239002.0, "step": 8075 }, { "entropy": 4.655025386810303, "epoch": 6.606952965235174, "grad_norm": 0.5859375, "learning_rate": 0.0001850407928979876, "loss": 4.2051, "mean_token_accuracy": 0.29307002425193784, "num_tokens": 131319418.0, "step": 8080 }, { "entropy": 4.652750158309937, "epoch": 6.611042944785276, "grad_norm": 0.57421875, "learning_rate": 0.00018475215464013828, "loss": 4.2159, "mean_token_accuracy": 0.2930891364812851, "num_tokens": 131400997.0, "step": 8085 }, { "entropy": 4.610401821136475, "epoch": 6.615132924335378, "grad_norm": 0.59375, "learning_rate": 0.00018446369326720975, "loss": 4.1807, "mean_token_accuracy": 0.29472084641456603, "num_tokens": 131481864.0, "step": 8090 }, { "entropy": 4.63852596282959, "epoch": 6.619222903885481, "grad_norm": 0.5703125, "learning_rate": 0.00018417540934458348, "loss": 4.1954, "mean_token_accuracy": 0.2928652256727219, "num_tokens": 131563554.0, "step": 8095 }, { "entropy": 4.61860876083374, "epoch": 6.623312883435583, "grad_norm": 0.5703125, "learning_rate": 0.00018388730343729331, "loss": 4.2012, "mean_token_accuracy": 0.2910107523202896, "num_tokens": 131644477.0, "step": 8100 }, { "entropy": 4.604886484146118, "epoch": 6.627402862985685, "grad_norm": 0.57421875, "learning_rate": 0.0001835993761100241, "loss": 4.1357, "mean_token_accuracy": 0.29756735563278197, "num_tokens": 131726104.0, "step": 8105 }, { "entropy": 4.621557283401489, "epoch": 6.6314928425357875, "grad_norm": 0.56640625, "learning_rate": 0.00018331162792711053, "loss": 4.2094, "mean_token_accuracy": 0.2933170735836029, "num_tokens": 131807781.0, "step": 8110 }, { "entropy": 4.646506118774414, "epoch": 6.6355828220858895, "grad_norm": 0.57421875, "learning_rate": 0.00018302405945253643, "loss": 4.1925, "mean_token_accuracy": 0.2944612979888916, "num_tokens": 131888792.0, "step": 8115 }, { "entropy": 4.6390900135040285, "epoch": 6.639672801635992, "grad_norm": 0.578125, "learning_rate": 0.00018273667124993315, "loss": 4.2144, "mean_token_accuracy": 0.2939594119787216, "num_tokens": 131970271.0, "step": 8120 }, { "entropy": 4.629734134674072, "epoch": 6.643762781186094, "grad_norm": 0.58203125, "learning_rate": 0.0001824494638825791, "loss": 4.2055, "mean_token_accuracy": 0.2940943717956543, "num_tokens": 132051830.0, "step": 8125 }, { "entropy": 4.688820457458496, "epoch": 6.647852760736196, "grad_norm": 0.5625, "learning_rate": 0.00018216243791339772, "loss": 4.2335, "mean_token_accuracy": 0.29244938790798186, "num_tokens": 132133290.0, "step": 8130 }, { "entropy": 4.6453454971313475, "epoch": 6.651942740286299, "grad_norm": 0.56640625, "learning_rate": 0.0001818755939049573, "loss": 4.2094, "mean_token_accuracy": 0.2921770513057709, "num_tokens": 132215013.0, "step": 8135 }, { "entropy": 4.624758720397949, "epoch": 6.656032719836401, "grad_norm": 0.57421875, "learning_rate": 0.00018158893241946943, "loss": 4.2175, "mean_token_accuracy": 0.29212296307086943, "num_tokens": 132295461.0, "step": 8140 }, { "entropy": 4.664038181304932, "epoch": 6.660122699386503, "grad_norm": 0.578125, "learning_rate": 0.0001813024540187878, "loss": 4.207, "mean_token_accuracy": 0.296927934885025, "num_tokens": 132377010.0, "step": 8145 }, { "entropy": 4.6528219223022464, "epoch": 6.664212678936606, "grad_norm": 0.59765625, "learning_rate": 0.00018101615926440745, "loss": 4.2031, "mean_token_accuracy": 0.2937388628721237, "num_tokens": 132458614.0, "step": 8150 }, { "entropy": 4.619529294967651, "epoch": 6.668302658486708, "grad_norm": 0.57421875, "learning_rate": 0.00018073004871746334, "loss": 4.1992, "mean_token_accuracy": 0.295290070772171, "num_tokens": 132539861.0, "step": 8155 }, { "entropy": 4.605418968200683, "epoch": 6.6723926380368095, "grad_norm": 0.5703125, "learning_rate": 0.0001804441229387294, "loss": 4.1798, "mean_token_accuracy": 0.29691497385501864, "num_tokens": 132621034.0, "step": 8160 }, { "entropy": 4.6036357402801515, "epoch": 6.676482617586912, "grad_norm": 0.5703125, "learning_rate": 0.00018015838248861744, "loss": 4.1749, "mean_token_accuracy": 0.29581704139709475, "num_tokens": 132701868.0, "step": 8165 }, { "entropy": 4.634261178970337, "epoch": 6.680572597137014, "grad_norm": 0.5703125, "learning_rate": 0.00017987282792717617, "loss": 4.1991, "mean_token_accuracy": 0.292342883348465, "num_tokens": 132783223.0, "step": 8170 }, { "entropy": 4.633958244323731, "epoch": 6.684662576687117, "grad_norm": 0.578125, "learning_rate": 0.0001795874598140897, "loss": 4.2054, "mean_token_accuracy": 0.29253250956535337, "num_tokens": 132864945.0, "step": 8175 }, { "entropy": 4.635308647155762, "epoch": 6.688752556237219, "grad_norm": 0.5703125, "learning_rate": 0.00017930227870867666, "loss": 4.1871, "mean_token_accuracy": 0.29319606721401215, "num_tokens": 132945186.0, "step": 8180 }, { "entropy": 4.650485324859619, "epoch": 6.692842535787321, "grad_norm": 0.5625, "learning_rate": 0.00017901728516988975, "loss": 4.2077, "mean_token_accuracy": 0.2935968697071075, "num_tokens": 133026811.0, "step": 8185 }, { "entropy": 4.636900949478149, "epoch": 6.696932515337423, "grad_norm": 0.58203125, "learning_rate": 0.00017873247975631312, "loss": 4.1873, "mean_token_accuracy": 0.294671568274498, "num_tokens": 133108413.0, "step": 8190 }, { "entropy": 4.595395755767822, "epoch": 6.701022494887526, "grad_norm": 0.55859375, "learning_rate": 0.000178447863026163, "loss": 4.1702, "mean_token_accuracy": 0.2975990414619446, "num_tokens": 133189298.0, "step": 8195 }, { "entropy": 4.569592571258545, "epoch": 6.705112474437628, "grad_norm": 0.5625, "learning_rate": 0.00017816343553728536, "loss": 4.1134, "mean_token_accuracy": 0.30393600165843965, "num_tokens": 133270875.0, "step": 8200 }, { "entropy": 4.583423137664795, "epoch": 6.70920245398773, "grad_norm": 0.58203125, "learning_rate": 0.00017787919784715544, "loss": 4.1407, "mean_token_accuracy": 0.2944240361452103, "num_tokens": 133351989.0, "step": 8205 }, { "entropy": 4.653213119506836, "epoch": 6.713292433537832, "grad_norm": 0.58203125, "learning_rate": 0.00017759515051287634, "loss": 4.2211, "mean_token_accuracy": 0.29180632531642914, "num_tokens": 133433619.0, "step": 8210 }, { "entropy": 4.60054383277893, "epoch": 6.717382413087934, "grad_norm": 0.56640625, "learning_rate": 0.00017731129409117823, "loss": 4.1449, "mean_token_accuracy": 0.2978470027446747, "num_tokens": 133514946.0, "step": 8215 }, { "entropy": 4.63750114440918, "epoch": 6.721472392638037, "grad_norm": 0.56640625, "learning_rate": 0.0001770276291384168, "loss": 4.2046, "mean_token_accuracy": 0.29350065886974336, "num_tokens": 133596605.0, "step": 8220 }, { "entropy": 4.652020359039307, "epoch": 6.725562372188139, "grad_norm": 0.58203125, "learning_rate": 0.00017674415621057304, "loss": 4.1837, "mean_token_accuracy": 0.29399036467075346, "num_tokens": 133677263.0, "step": 8225 }, { "entropy": 4.6225381851196286, "epoch": 6.729652351738241, "grad_norm": 0.58984375, "learning_rate": 0.00017646087586325088, "loss": 4.1867, "mean_token_accuracy": 0.29936026930809023, "num_tokens": 133757543.0, "step": 8230 }, { "entropy": 4.594828939437866, "epoch": 6.733742331288344, "grad_norm": 0.5703125, "learning_rate": 0.0001761777886516771, "loss": 4.1692, "mean_token_accuracy": 0.2960600972175598, "num_tokens": 133838988.0, "step": 8235 }, { "entropy": 4.648728513717652, "epoch": 6.737832310838446, "grad_norm": 0.5703125, "learning_rate": 0.00017589489513070028, "loss": 4.194, "mean_token_accuracy": 0.294809028506279, "num_tokens": 133920592.0, "step": 8240 }, { "entropy": 4.6253008365631105, "epoch": 6.741922290388548, "grad_norm": 0.56640625, "learning_rate": 0.00017561219585478862, "loss": 4.1829, "mean_token_accuracy": 0.2943107008934021, "num_tokens": 134002284.0, "step": 8245 }, { "entropy": 4.616927289962769, "epoch": 6.74601226993865, "grad_norm": 0.57421875, "learning_rate": 0.0001753296913780303, "loss": 4.192, "mean_token_accuracy": 0.29319021105766296, "num_tokens": 134083639.0, "step": 8250 }, { "entropy": 4.599713134765625, "epoch": 6.750102249488752, "grad_norm": 0.5703125, "learning_rate": 0.0001750473822541313, "loss": 4.1445, "mean_token_accuracy": 0.2978177934885025, "num_tokens": 134165251.0, "step": 8255 }, { "entropy": 4.628828525543213, "epoch": 6.754192229038855, "grad_norm": 0.57421875, "learning_rate": 0.00017476526903641488, "loss": 4.2051, "mean_token_accuracy": 0.2932108938694, "num_tokens": 134246161.0, "step": 8260 }, { "entropy": 4.645445728302002, "epoch": 6.758282208588957, "grad_norm": 0.578125, "learning_rate": 0.00017448335227782016, "loss": 4.2181, "mean_token_accuracy": 0.2933057636022568, "num_tokens": 134327226.0, "step": 8265 }, { "entropy": 4.631940221786499, "epoch": 6.762372188139059, "grad_norm": 0.5625, "learning_rate": 0.0001742016325309016, "loss": 4.1959, "mean_token_accuracy": 0.2934738039970398, "num_tokens": 134408861.0, "step": 8270 }, { "entropy": 4.663918542861938, "epoch": 6.766462167689162, "grad_norm": 0.5625, "learning_rate": 0.0001739201103478269, "loss": 4.2255, "mean_token_accuracy": 0.29073479771614075, "num_tokens": 134490377.0, "step": 8275 }, { "entropy": 4.620434904098511, "epoch": 6.770552147239264, "grad_norm": 0.5703125, "learning_rate": 0.00017363878628037693, "loss": 4.1807, "mean_token_accuracy": 0.29572502970695497, "num_tokens": 134571108.0, "step": 8280 }, { "entropy": 4.599979448318481, "epoch": 6.774642126789366, "grad_norm": 0.5625, "learning_rate": 0.00017335766087994443, "loss": 4.1653, "mean_token_accuracy": 0.2967634916305542, "num_tokens": 134652326.0, "step": 8285 }, { "entropy": 4.644181108474731, "epoch": 6.7787321063394685, "grad_norm": 0.56640625, "learning_rate": 0.00017307673469753215, "loss": 4.2239, "mean_token_accuracy": 0.2948341935873032, "num_tokens": 134733298.0, "step": 8290 }, { "entropy": 4.649595260620117, "epoch": 6.7828220858895705, "grad_norm": 0.58984375, "learning_rate": 0.00017279600828375304, "loss": 4.2394, "mean_token_accuracy": 0.28875263333320617, "num_tokens": 134814897.0, "step": 8295 }, { "entropy": 4.622741603851319, "epoch": 6.786912065439672, "grad_norm": 0.5625, "learning_rate": 0.00017251548218882813, "loss": 4.1564, "mean_token_accuracy": 0.296170175075531, "num_tokens": 134896493.0, "step": 8300 }, { "entropy": 4.667420387268066, "epoch": 6.791002044989775, "grad_norm": 0.5859375, "learning_rate": 0.0001722351569625859, "loss": 4.2489, "mean_token_accuracy": 0.2910971999168396, "num_tokens": 134977255.0, "step": 8305 }, { "entropy": 4.617430210113525, "epoch": 6.795092024539877, "grad_norm": 0.56640625, "learning_rate": 0.00017195503315446109, "loss": 4.2003, "mean_token_accuracy": 0.2911302983760834, "num_tokens": 135058914.0, "step": 8310 }, { "entropy": 4.676914930343628, "epoch": 6.79918200408998, "grad_norm": 0.578125, "learning_rate": 0.00017167511131349373, "loss": 4.2351, "mean_token_accuracy": 0.29191267788410186, "num_tokens": 135140117.0, "step": 8315 }, { "entropy": 4.617963790893555, "epoch": 6.803271983640082, "grad_norm": 0.5625, "learning_rate": 0.000171395391988328, "loss": 4.1744, "mean_token_accuracy": 0.2952577292919159, "num_tokens": 135221553.0, "step": 8320 }, { "entropy": 4.624241495132447, "epoch": 6.807361963190184, "grad_norm": 0.578125, "learning_rate": 0.00017111587572721105, "loss": 4.2055, "mean_token_accuracy": 0.2897685021162033, "num_tokens": 135302612.0, "step": 8325 }, { "entropy": 4.626815748214722, "epoch": 6.811451942740287, "grad_norm": 0.58203125, "learning_rate": 0.0001708365630779922, "loss": 4.1906, "mean_token_accuracy": 0.2967037230730057, "num_tokens": 135384046.0, "step": 8330 }, { "entropy": 4.625767755508423, "epoch": 6.815541922290389, "grad_norm": 0.56640625, "learning_rate": 0.00017055745458812143, "loss": 4.1846, "mean_token_accuracy": 0.29507235884666444, "num_tokens": 135464833.0, "step": 8335 }, { "entropy": 4.620637893676758, "epoch": 6.8196319018404905, "grad_norm": 0.56640625, "learning_rate": 0.00017027855080464904, "loss": 4.1946, "mean_token_accuracy": 0.294300577044487, "num_tokens": 135546428.0, "step": 8340 }, { "entropy": 4.622606945037842, "epoch": 6.823721881390593, "grad_norm": 0.578125, "learning_rate": 0.00016999985227422344, "loss": 4.1868, "mean_token_accuracy": 0.29448942840099335, "num_tokens": 135627628.0, "step": 8345 }, { "entropy": 4.650513982772827, "epoch": 6.827811860940695, "grad_norm": 0.56640625, "learning_rate": 0.0001697213595430913, "loss": 4.2158, "mean_token_accuracy": 0.292169389128685, "num_tokens": 135709185.0, "step": 8350 }, { "entropy": 4.60190863609314, "epoch": 6.831901840490797, "grad_norm": 0.5625, "learning_rate": 0.00016944307315709575, "loss": 4.1495, "mean_token_accuracy": 0.2965735226869583, "num_tokens": 135790370.0, "step": 8355 }, { "entropy": 4.616092538833618, "epoch": 6.8359918200409, "grad_norm": 0.5703125, "learning_rate": 0.00016916499366167542, "loss": 4.1905, "mean_token_accuracy": 0.29786417782306673, "num_tokens": 135871786.0, "step": 8360 }, { "entropy": 4.645244550704956, "epoch": 6.840081799591002, "grad_norm": 0.578125, "learning_rate": 0.00016888712160186337, "loss": 4.1938, "mean_token_accuracy": 0.2938934713602066, "num_tokens": 135953457.0, "step": 8365 }, { "entropy": 4.632488012313843, "epoch": 6.844171779141105, "grad_norm": 0.57421875, "learning_rate": 0.0001686094575222865, "loss": 4.2106, "mean_token_accuracy": 0.29379266798496245, "num_tokens": 136035111.0, "step": 8370 }, { "entropy": 4.634761142730713, "epoch": 6.848261758691207, "grad_norm": 0.5703125, "learning_rate": 0.00016833200196716353, "loss": 4.1885, "mean_token_accuracy": 0.2932149887084961, "num_tokens": 136116361.0, "step": 8375 }, { "entropy": 4.644320106506347, "epoch": 6.852351738241309, "grad_norm": 0.58203125, "learning_rate": 0.00016805475548030464, "loss": 4.2069, "mean_token_accuracy": 0.29588742554187775, "num_tokens": 136197654.0, "step": 8380 }, { "entropy": 4.5951817512512205, "epoch": 6.856441717791411, "grad_norm": 0.55859375, "learning_rate": 0.00016777771860511061, "loss": 4.1673, "mean_token_accuracy": 0.29768737256526945, "num_tokens": 136279288.0, "step": 8385 }, { "entropy": 4.6224973678588865, "epoch": 6.860531697341513, "grad_norm": 0.5703125, "learning_rate": 0.0001675008918845708, "loss": 4.197, "mean_token_accuracy": 0.2959954023361206, "num_tokens": 136360998.0, "step": 8390 }, { "entropy": 4.668500137329102, "epoch": 6.864621676891615, "grad_norm": 0.578125, "learning_rate": 0.0001672242758612631, "loss": 4.21, "mean_token_accuracy": 0.29333322644233706, "num_tokens": 136442595.0, "step": 8395 }, { "entropy": 4.609576320648193, "epoch": 6.868711656441718, "grad_norm": 0.5546875, "learning_rate": 0.0001669478710773523, "loss": 4.1643, "mean_token_accuracy": 0.2949462622404099, "num_tokens": 136523714.0, "step": 8400 }, { "entropy": 4.614280414581299, "epoch": 6.87280163599182, "grad_norm": 0.58203125, "learning_rate": 0.00016667167807458917, "loss": 4.1669, "mean_token_accuracy": 0.30128417909145355, "num_tokens": 136604516.0, "step": 8405 }, { "entropy": 4.5856725692749025, "epoch": 6.876891615541922, "grad_norm": 0.56640625, "learning_rate": 0.00016639569739430928, "loss": 4.1355, "mean_token_accuracy": 0.2998155266046524, "num_tokens": 136685655.0, "step": 8410 }, { "entropy": 4.6330952644348145, "epoch": 6.880981595092025, "grad_norm": 0.55859375, "learning_rate": 0.00016611992957743217, "loss": 4.163, "mean_token_accuracy": 0.2982346534729004, "num_tokens": 136767343.0, "step": 8415 }, { "entropy": 4.591008615493775, "epoch": 6.885071574642127, "grad_norm": 0.58984375, "learning_rate": 0.00016584437516446021, "loss": 4.1415, "mean_token_accuracy": 0.295902743935585, "num_tokens": 136848675.0, "step": 8420 }, { "entropy": 4.599712991714478, "epoch": 6.8891615541922295, "grad_norm": 0.55859375, "learning_rate": 0.00016556903469547738, "loss": 4.1785, "mean_token_accuracy": 0.29519915878772734, "num_tokens": 136930023.0, "step": 8425 }, { "entropy": 4.690613174438477, "epoch": 6.8932515337423315, "grad_norm": 0.578125, "learning_rate": 0.0001652939087101484, "loss": 4.2549, "mean_token_accuracy": 0.28928124010562895, "num_tokens": 137011237.0, "step": 8430 }, { "entropy": 4.62441120147705, "epoch": 6.897341513292433, "grad_norm": 0.5703125, "learning_rate": 0.0001650189977477175, "loss": 4.2169, "mean_token_accuracy": 0.2909060835838318, "num_tokens": 137092826.0, "step": 8435 }, { "entropy": 4.646114301681519, "epoch": 6.901431492842535, "grad_norm": 0.56640625, "learning_rate": 0.00016474430234700784, "loss": 4.2211, "mean_token_accuracy": 0.29264298379421233, "num_tokens": 137174425.0, "step": 8440 }, { "entropy": 4.641643953323364, "epoch": 6.905521472392638, "grad_norm": 0.58203125, "learning_rate": 0.00016446982304641944, "loss": 4.1834, "mean_token_accuracy": 0.2986262619495392, "num_tokens": 137256052.0, "step": 8445 }, { "entropy": 4.615571403503418, "epoch": 6.90961145194274, "grad_norm": 0.56640625, "learning_rate": 0.00016419556038392935, "loss": 4.1978, "mean_token_accuracy": 0.2928495556116104, "num_tokens": 137337740.0, "step": 8450 }, { "entropy": 4.619269323348999, "epoch": 6.913701431492843, "grad_norm": 0.56640625, "learning_rate": 0.0001639215148970898, "loss": 4.197, "mean_token_accuracy": 0.29453299641609193, "num_tokens": 137419286.0, "step": 8455 }, { "entropy": 4.630918645858765, "epoch": 6.917791411042945, "grad_norm": 0.56640625, "learning_rate": 0.00016364768712302718, "loss": 4.1946, "mean_token_accuracy": 0.2967101216316223, "num_tokens": 137500363.0, "step": 8460 }, { "entropy": 4.611219072341919, "epoch": 6.921881390593047, "grad_norm": 0.56640625, "learning_rate": 0.00016337407759844146, "loss": 4.1838, "mean_token_accuracy": 0.29355780482292176, "num_tokens": 137581954.0, "step": 8465 }, { "entropy": 4.6445049285888675, "epoch": 6.92597137014315, "grad_norm": 0.58203125, "learning_rate": 0.0001631006868596047, "loss": 4.212, "mean_token_accuracy": 0.29581980407238007, "num_tokens": 137663109.0, "step": 8470 }, { "entropy": 4.60771336555481, "epoch": 6.9300613496932515, "grad_norm": 0.56640625, "learning_rate": 0.00016282751544236012, "loss": 4.1611, "mean_token_accuracy": 0.2988464266061783, "num_tokens": 137744531.0, "step": 8475 }, { "entropy": 4.613905429840088, "epoch": 6.934151329243353, "grad_norm": 0.5625, "learning_rate": 0.00016255456388212107, "loss": 4.1796, "mean_token_accuracy": 0.2955861955881119, "num_tokens": 137826138.0, "step": 8480 }, { "entropy": 4.6367240905761715, "epoch": 6.938241308793456, "grad_norm": 0.5546875, "learning_rate": 0.0001622818327138702, "loss": 4.1929, "mean_token_accuracy": 0.2951211631298065, "num_tokens": 137907782.0, "step": 8485 }, { "entropy": 4.645176362991333, "epoch": 6.942331288343558, "grad_norm": 0.5859375, "learning_rate": 0.00016200932247215771, "loss": 4.1979, "mean_token_accuracy": 0.2967901825904846, "num_tokens": 137988813.0, "step": 8490 }, { "entropy": 4.62744026184082, "epoch": 6.94642126789366, "grad_norm": 0.57421875, "learning_rate": 0.00016173703369110138, "loss": 4.2069, "mean_token_accuracy": 0.2944197982549667, "num_tokens": 138070371.0, "step": 8495 }, { "entropy": 4.57274751663208, "epoch": 6.950511247443763, "grad_norm": 0.58203125, "learning_rate": 0.00016146496690438446, "loss": 4.1512, "mean_token_accuracy": 0.2976902574300766, "num_tokens": 138150724.0, "step": 8500 }, { "epoch": 6.950511247443763, "eval_entropy": 4.448177619330211, "eval_loss": 4.58236837387085, "eval_mean_token_accuracy": 0.27018642293210193, "eval_num_tokens": 138150724.0, "eval_runtime": 65.3799, "eval_samples_per_second": 63.276, "eval_steps_per_second": 7.923, "step": 8500 }, { "entropy": 4.6109456539154055, "epoch": 6.954601226993865, "grad_norm": 0.57421875, "learning_rate": 0.00016119312264525537, "loss": 4.1896, "mean_token_accuracy": 0.2972770780324936, "num_tokens": 138232302.0, "step": 8505 }, { "entropy": 4.618128490447998, "epoch": 6.958691206543968, "grad_norm": 0.58203125, "learning_rate": 0.00016092150144652622, "loss": 4.1671, "mean_token_accuracy": 0.29874907433986664, "num_tokens": 138313860.0, "step": 8510 }, { "entropy": 4.625085401535034, "epoch": 6.96278118609407, "grad_norm": 0.58203125, "learning_rate": 0.000160650103840572, "loss": 4.1751, "mean_token_accuracy": 0.29663399457931516, "num_tokens": 138395357.0, "step": 8515 }, { "entropy": 4.637985801696777, "epoch": 6.9668711656441715, "grad_norm": 0.57421875, "learning_rate": 0.00016037893035932943, "loss": 4.2106, "mean_token_accuracy": 0.2925273895263672, "num_tokens": 138476930.0, "step": 8520 }, { "entropy": 4.615249872207642, "epoch": 6.970961145194274, "grad_norm": 0.578125, "learning_rate": 0.00016010798153429597, "loss": 4.1841, "mean_token_accuracy": 0.29653716683387754, "num_tokens": 138558300.0, "step": 8525 }, { "entropy": 4.580342245101929, "epoch": 6.975051124744376, "grad_norm": 0.58984375, "learning_rate": 0.00015983725789652878, "loss": 4.1435, "mean_token_accuracy": 0.29809763133525846, "num_tokens": 138639951.0, "step": 8530 }, { "entropy": 4.6176838874816895, "epoch": 6.979141104294478, "grad_norm": 0.56640625, "learning_rate": 0.0001595667599766435, "loss": 4.1766, "mean_token_accuracy": 0.29680508375167847, "num_tokens": 138721507.0, "step": 8535 }, { "entropy": 4.708914947509766, "epoch": 6.983231083844581, "grad_norm": 0.58984375, "learning_rate": 0.00015929648830481375, "loss": 4.2777, "mean_token_accuracy": 0.28579892218112946, "num_tokens": 138802506.0, "step": 8540 }, { "entropy": 4.616736316680909, "epoch": 6.987321063394683, "grad_norm": 0.57421875, "learning_rate": 0.00015902644341076909, "loss": 4.159, "mean_token_accuracy": 0.29839507639408114, "num_tokens": 138884103.0, "step": 8545 }, { "entropy": 4.656926870346069, "epoch": 6.991411042944785, "grad_norm": 0.59375, "learning_rate": 0.00015875662582379518, "loss": 4.2363, "mean_token_accuracy": 0.2898046255111694, "num_tokens": 138965392.0, "step": 8550 }, { "entropy": 4.635281562805176, "epoch": 6.995501022494888, "grad_norm": 0.5703125, "learning_rate": 0.00015848703607273195, "loss": 4.1987, "mean_token_accuracy": 0.29509333074092864, "num_tokens": 139046385.0, "step": 8555 }, { "entropy": 4.662484741210937, "epoch": 6.99959100204499, "grad_norm": 0.57421875, "learning_rate": 0.00015821767468597264, "loss": 4.2368, "mean_token_accuracy": 0.2912438571453094, "num_tokens": 139127512.0, "step": 8560 }, { "entropy": 4.682872560289171, "epoch": 7.003271983640082, "grad_norm": 0.59375, "learning_rate": 0.0001579485421914631, "loss": 4.2092, "mean_token_accuracy": 0.29650963015026516, "num_tokens": 139192621.0, "step": 8565 }, { "entropy": 4.605118751525879, "epoch": 7.007361963190184, "grad_norm": 0.57421875, "learning_rate": 0.00015767963911670038, "loss": 4.0972, "mean_token_accuracy": 0.30177919268608094, "num_tokens": 139273790.0, "step": 8570 }, { "entropy": 4.5931449890136715, "epoch": 7.011451942740286, "grad_norm": 0.57421875, "learning_rate": 0.000157410965988732, "loss": 4.0693, "mean_token_accuracy": 0.3063156843185425, "num_tokens": 139355182.0, "step": 8575 }, { "entropy": 4.613659858703613, "epoch": 7.015541922290389, "grad_norm": 0.58203125, "learning_rate": 0.0001571425233341547, "loss": 4.0907, "mean_token_accuracy": 0.3026617616415024, "num_tokens": 139436806.0, "step": 8580 }, { "entropy": 4.589653491973877, "epoch": 7.019631901840491, "grad_norm": 0.5703125, "learning_rate": 0.00015687431167911371, "loss": 4.0824, "mean_token_accuracy": 0.30186585783958436, "num_tokens": 139517896.0, "step": 8585 }, { "entropy": 4.6332597732543945, "epoch": 7.023721881390593, "grad_norm": 0.59765625, "learning_rate": 0.00015660633154930103, "loss": 4.108, "mean_token_accuracy": 0.3022927165031433, "num_tokens": 139599447.0, "step": 8590 }, { "entropy": 4.604008531570434, "epoch": 7.027811860940695, "grad_norm": 0.578125, "learning_rate": 0.00015633858346995544, "loss": 4.0929, "mean_token_accuracy": 0.3067808270454407, "num_tokens": 139680690.0, "step": 8595 }, { "entropy": 4.622136354446411, "epoch": 7.031901840490797, "grad_norm": 0.57421875, "learning_rate": 0.00015607106796586056, "loss": 4.1113, "mean_token_accuracy": 0.30823162496089934, "num_tokens": 139761634.0, "step": 8600 }, { "entropy": 4.625117397308349, "epoch": 7.0359918200409, "grad_norm": 0.59375, "learning_rate": 0.0001558037855613442, "loss": 4.1072, "mean_token_accuracy": 0.3014308840036392, "num_tokens": 139842927.0, "step": 8605 }, { "entropy": 4.6121275424957275, "epoch": 7.040081799591002, "grad_norm": 0.59375, "learning_rate": 0.00015553673678027734, "loss": 4.1056, "mean_token_accuracy": 0.3043176710605621, "num_tokens": 139924217.0, "step": 8610 }, { "entropy": 4.60748782157898, "epoch": 7.044171779141104, "grad_norm": 0.5859375, "learning_rate": 0.00015526992214607308, "loss": 4.0961, "mean_token_accuracy": 0.30656982958316803, "num_tokens": 140005470.0, "step": 8615 }, { "entropy": 4.668444585800171, "epoch": 7.048261758691207, "grad_norm": 0.58984375, "learning_rate": 0.00015500334218168554, "loss": 4.1499, "mean_token_accuracy": 0.30209210216999055, "num_tokens": 140086739.0, "step": 8620 }, { "entropy": 4.6035431861877445, "epoch": 7.052351738241309, "grad_norm": 0.5859375, "learning_rate": 0.000154736997409609, "loss": 4.0874, "mean_token_accuracy": 0.30579097867012023, "num_tokens": 140168364.0, "step": 8625 }, { "entropy": 4.621374320983887, "epoch": 7.056441717791411, "grad_norm": 0.6015625, "learning_rate": 0.00015447088835187655, "loss": 4.1235, "mean_token_accuracy": 0.30124713480472565, "num_tokens": 140249948.0, "step": 8630 }, { "entropy": 4.5881226539611815, "epoch": 7.0605316973415135, "grad_norm": 0.5859375, "learning_rate": 0.0001542050155300595, "loss": 4.077, "mean_token_accuracy": 0.30446280241012574, "num_tokens": 140331246.0, "step": 8635 }, { "entropy": 4.6275214672088625, "epoch": 7.0646216768916155, "grad_norm": 0.58984375, "learning_rate": 0.0001539393794652662, "loss": 4.1122, "mean_token_accuracy": 0.30213204622268675, "num_tokens": 140412474.0, "step": 8640 }, { "entropy": 4.595401430130005, "epoch": 7.068711656441717, "grad_norm": 0.59375, "learning_rate": 0.00015367398067814048, "loss": 4.0922, "mean_token_accuracy": 0.3017816483974457, "num_tokens": 140493886.0, "step": 8645 }, { "entropy": 4.568799448013306, "epoch": 7.07280163599182, "grad_norm": 0.58984375, "learning_rate": 0.00015340881968886171, "loss": 4.0531, "mean_token_accuracy": 0.3056829422712326, "num_tokens": 140574357.0, "step": 8650 }, { "entropy": 4.594132900238037, "epoch": 7.076891615541922, "grad_norm": 0.578125, "learning_rate": 0.00015314389701714286, "loss": 4.0798, "mean_token_accuracy": 0.30474735200405123, "num_tokens": 140656048.0, "step": 8655 }, { "entropy": 4.631674146652221, "epoch": 7.080981595092025, "grad_norm": 0.59375, "learning_rate": 0.00015287921318222987, "loss": 4.1163, "mean_token_accuracy": 0.302283388376236, "num_tokens": 140737714.0, "step": 8660 }, { "entropy": 4.620348310470581, "epoch": 7.085071574642127, "grad_norm": 0.578125, "learning_rate": 0.00015261476870290049, "loss": 4.1263, "mean_token_accuracy": 0.3022204041481018, "num_tokens": 140818959.0, "step": 8665 }, { "entropy": 4.582056331634521, "epoch": 7.089161554192229, "grad_norm": 0.58984375, "learning_rate": 0.0001523505640974634, "loss": 4.0756, "mean_token_accuracy": 0.303360190987587, "num_tokens": 140900133.0, "step": 8670 }, { "entropy": 4.614884948730468, "epoch": 7.093251533742332, "grad_norm": 0.609375, "learning_rate": 0.00015208659988375715, "loss": 4.1037, "mean_token_accuracy": 0.3007459223270416, "num_tokens": 140981824.0, "step": 8675 }, { "entropy": 4.63407711982727, "epoch": 7.097341513292434, "grad_norm": 0.58984375, "learning_rate": 0.000151822876579149, "loss": 4.1336, "mean_token_accuracy": 0.3006787359714508, "num_tokens": 141063179.0, "step": 8680 }, { "entropy": 4.586604642868042, "epoch": 7.1014314928425355, "grad_norm": 0.5859375, "learning_rate": 0.0001515593947005343, "loss": 4.0684, "mean_token_accuracy": 0.3058437407016754, "num_tokens": 141144402.0, "step": 8685 }, { "entropy": 4.533650684356689, "epoch": 7.105521472392638, "grad_norm": 0.609375, "learning_rate": 0.0001512961547643349, "loss": 4.0238, "mean_token_accuracy": 0.31228741705417634, "num_tokens": 141225752.0, "step": 8690 }, { "entropy": 4.589834833145142, "epoch": 7.10961145194274, "grad_norm": 0.59765625, "learning_rate": 0.00015103315728649857, "loss": 4.0923, "mean_token_accuracy": 0.30627498030662537, "num_tokens": 141306930.0, "step": 8695 }, { "entropy": 4.5792093753814695, "epoch": 7.113701431492842, "grad_norm": 0.58203125, "learning_rate": 0.000150770402782498, "loss": 4.0778, "mean_token_accuracy": 0.30604792237281797, "num_tokens": 141388540.0, "step": 8700 }, { "entropy": 4.651524591445923, "epoch": 7.117791411042945, "grad_norm": 0.5859375, "learning_rate": 0.0001505078917673294, "loss": 4.1386, "mean_token_accuracy": 0.3015183866024017, "num_tokens": 141470127.0, "step": 8705 }, { "entropy": 4.616889476776123, "epoch": 7.121881390593047, "grad_norm": 0.59375, "learning_rate": 0.0001502456247555119, "loss": 4.0955, "mean_token_accuracy": 0.3043801635503769, "num_tokens": 141550789.0, "step": 8710 }, { "entropy": 4.601114416122437, "epoch": 7.12597137014315, "grad_norm": 0.59375, "learning_rate": 0.00014998360226108637, "loss": 4.0781, "mean_token_accuracy": 0.305596786737442, "num_tokens": 141632413.0, "step": 8715 }, { "entropy": 4.590821313858032, "epoch": 7.130061349693252, "grad_norm": 0.60546875, "learning_rate": 0.00014972182479761442, "loss": 4.0936, "mean_token_accuracy": 0.3039992868900299, "num_tokens": 141713293.0, "step": 8720 }, { "entropy": 4.596924304962158, "epoch": 7.134151329243354, "grad_norm": 0.58203125, "learning_rate": 0.00014946029287817735, "loss": 4.1017, "mean_token_accuracy": 0.3058774799108505, "num_tokens": 141794861.0, "step": 8725 }, { "entropy": 4.633793544769287, "epoch": 7.138241308793456, "grad_norm": 0.59375, "learning_rate": 0.00014919900701537535, "loss": 4.127, "mean_token_accuracy": 0.3035710871219635, "num_tokens": 141876008.0, "step": 8730 }, { "entropy": 4.621849870681762, "epoch": 7.142331288343558, "grad_norm": 0.59375, "learning_rate": 0.000148937967721326, "loss": 4.0994, "mean_token_accuracy": 0.30480864346027375, "num_tokens": 141957424.0, "step": 8735 }, { "entropy": 4.607697057723999, "epoch": 7.14642126789366, "grad_norm": 0.60546875, "learning_rate": 0.00014867717550766417, "loss": 4.0831, "mean_token_accuracy": 0.3040384531021118, "num_tokens": 142038640.0, "step": 8740 }, { "entropy": 4.570908069610596, "epoch": 7.150511247443763, "grad_norm": 0.59375, "learning_rate": 0.0001484166308855398, "loss": 4.061, "mean_token_accuracy": 0.3071552574634552, "num_tokens": 142119754.0, "step": 8745 }, { "entropy": 4.622466659545898, "epoch": 7.154601226993865, "grad_norm": 0.609375, "learning_rate": 0.00014815633436561816, "loss": 4.1371, "mean_token_accuracy": 0.2994953364133835, "num_tokens": 142200250.0, "step": 8750 }, { "entropy": 4.606250524520874, "epoch": 7.158691206543967, "grad_norm": 0.59765625, "learning_rate": 0.00014789628645807773, "loss": 4.1286, "mean_token_accuracy": 0.30149418115615845, "num_tokens": 142281096.0, "step": 8755 }, { "entropy": 4.5990842342376705, "epoch": 7.16278118609407, "grad_norm": 0.59765625, "learning_rate": 0.0001476364876726101, "loss": 4.1076, "mean_token_accuracy": 0.30342867970466614, "num_tokens": 142362764.0, "step": 8760 }, { "entropy": 4.6476318359375, "epoch": 7.166871165644172, "grad_norm": 0.59375, "learning_rate": 0.00014737693851841834, "loss": 4.1267, "mean_token_accuracy": 0.3008619427680969, "num_tokens": 142444419.0, "step": 8765 }, { "entropy": 4.610910320281983, "epoch": 7.170961145194274, "grad_norm": 0.59375, "learning_rate": 0.00014711763950421637, "loss": 4.1016, "mean_token_accuracy": 0.30667617321014407, "num_tokens": 142525712.0, "step": 8770 }, { "entropy": 4.595026874542237, "epoch": 7.1750511247443765, "grad_norm": 0.625, "learning_rate": 0.0001468585911382278, "loss": 4.0966, "mean_token_accuracy": 0.3045433282852173, "num_tokens": 142607105.0, "step": 8775 }, { "entropy": 4.619439220428466, "epoch": 7.179141104294478, "grad_norm": 0.578125, "learning_rate": 0.00014659979392818486, "loss": 4.1015, "mean_token_accuracy": 0.30413382053375243, "num_tokens": 142688652.0, "step": 8780 }, { "entropy": 4.628879070281982, "epoch": 7.183231083844581, "grad_norm": 0.578125, "learning_rate": 0.00014634124838132782, "loss": 4.1296, "mean_token_accuracy": 0.30323927402496337, "num_tokens": 142770031.0, "step": 8785 }, { "entropy": 4.590967893600464, "epoch": 7.187321063394683, "grad_norm": 0.59765625, "learning_rate": 0.0001460829550044033, "loss": 4.1119, "mean_token_accuracy": 0.3006419539451599, "num_tokens": 142851252.0, "step": 8790 }, { "entropy": 4.632367038726807, "epoch": 7.191411042944785, "grad_norm": 0.58984375, "learning_rate": 0.00014582491430366402, "loss": 4.1334, "mean_token_accuracy": 0.30546108484268186, "num_tokens": 142932133.0, "step": 8795 }, { "entropy": 4.593401670455933, "epoch": 7.195501022494888, "grad_norm": 0.59765625, "learning_rate": 0.00014556712678486728, "loss": 4.0913, "mean_token_accuracy": 0.30466486513614655, "num_tokens": 143013383.0, "step": 8800 }, { "entropy": 4.588053131103516, "epoch": 7.19959100204499, "grad_norm": 0.59375, "learning_rate": 0.00014530959295327411, "loss": 4.0899, "mean_token_accuracy": 0.30237488746643065, "num_tokens": 143094157.0, "step": 8805 }, { "entropy": 4.629646873474121, "epoch": 7.203680981595092, "grad_norm": 0.6015625, "learning_rate": 0.00014505231331364844, "loss": 4.1307, "mean_token_accuracy": 0.30172920525074004, "num_tokens": 143175052.0, "step": 8810 }, { "entropy": 4.6406943798065186, "epoch": 7.2077709611451946, "grad_norm": 0.609375, "learning_rate": 0.00014479528837025584, "loss": 4.1477, "mean_token_accuracy": 0.29836204648017883, "num_tokens": 143255607.0, "step": 8815 }, { "entropy": 4.601746845245361, "epoch": 7.2118609406952965, "grad_norm": 0.5859375, "learning_rate": 0.00014453851862686288, "loss": 4.1152, "mean_token_accuracy": 0.3028663367033005, "num_tokens": 143337239.0, "step": 8820 }, { "entropy": 4.631759071350098, "epoch": 7.215950920245398, "grad_norm": 0.58203125, "learning_rate": 0.00014428200458673572, "loss": 4.1401, "mean_token_accuracy": 0.2999334692955017, "num_tokens": 143418492.0, "step": 8825 }, { "entropy": 4.643289804458618, "epoch": 7.220040899795501, "grad_norm": 0.60546875, "learning_rate": 0.00014402574675263945, "loss": 4.1657, "mean_token_accuracy": 0.29839341044425965, "num_tokens": 143500086.0, "step": 8830 }, { "entropy": 4.631990623474121, "epoch": 7.224130879345603, "grad_norm": 0.61328125, "learning_rate": 0.000143769745626837, "loss": 4.1242, "mean_token_accuracy": 0.302613827586174, "num_tokens": 143581181.0, "step": 8835 }, { "entropy": 4.622196674346924, "epoch": 7.228220858895705, "grad_norm": 0.5859375, "learning_rate": 0.0001435140017110883, "loss": 4.0991, "mean_token_accuracy": 0.3032923758029938, "num_tokens": 143662773.0, "step": 8840 }, { "entropy": 4.556346988677978, "epoch": 7.232310838445808, "grad_norm": 0.609375, "learning_rate": 0.0001432585155066487, "loss": 4.0707, "mean_token_accuracy": 0.30518904626369475, "num_tokens": 143743580.0, "step": 8845 }, { "entropy": 4.598207712173462, "epoch": 7.23640081799591, "grad_norm": 0.59765625, "learning_rate": 0.00014300328751426904, "loss": 4.0963, "mean_token_accuracy": 0.3070404648780823, "num_tokens": 143824308.0, "step": 8850 }, { "entropy": 4.639033699035645, "epoch": 7.240490797546013, "grad_norm": 0.578125, "learning_rate": 0.0001427483182341936, "loss": 4.1231, "mean_token_accuracy": 0.30022102892398833, "num_tokens": 143905943.0, "step": 8855 }, { "entropy": 4.602065944671631, "epoch": 7.244580777096115, "grad_norm": 0.58984375, "learning_rate": 0.00014249360816615976, "loss": 4.118, "mean_token_accuracy": 0.3015020668506622, "num_tokens": 143986857.0, "step": 8860 }, { "entropy": 4.5389204025268555, "epoch": 7.2486707566462165, "grad_norm": 0.59375, "learning_rate": 0.00014223915780939685, "loss": 4.051, "mean_token_accuracy": 0.3077487647533417, "num_tokens": 144068137.0, "step": 8865 }, { "entropy": 4.598529720306397, "epoch": 7.252760736196319, "grad_norm": 0.58984375, "learning_rate": 0.0001419849676626252, "loss": 4.1227, "mean_token_accuracy": 0.3033046692609787, "num_tokens": 144149546.0, "step": 8870 }, { "entropy": 4.573176622390747, "epoch": 7.256850715746421, "grad_norm": 0.58984375, "learning_rate": 0.00014173103822405503, "loss": 4.0482, "mean_token_accuracy": 0.3093693882226944, "num_tokens": 144230828.0, "step": 8875 }, { "entropy": 4.582588338851929, "epoch": 7.260940695296523, "grad_norm": 0.5859375, "learning_rate": 0.00014147736999138555, "loss": 4.0967, "mean_token_accuracy": 0.3039365172386169, "num_tokens": 144312362.0, "step": 8880 }, { "entropy": 4.585626697540283, "epoch": 7.265030674846626, "grad_norm": 0.59375, "learning_rate": 0.0001412239634618044, "loss": 4.1239, "mean_token_accuracy": 0.3033891707658768, "num_tokens": 144393979.0, "step": 8885 }, { "entropy": 4.595276975631714, "epoch": 7.269120654396728, "grad_norm": 0.60546875, "learning_rate": 0.00014097081913198555, "loss": 4.0995, "mean_token_accuracy": 0.3025997430086136, "num_tokens": 144475635.0, "step": 8890 }, { "entropy": 4.650474929809571, "epoch": 7.27321063394683, "grad_norm": 0.60546875, "learning_rate": 0.00014071793749808982, "loss": 4.1615, "mean_token_accuracy": 0.2980715185403824, "num_tokens": 144557243.0, "step": 8895 }, { "entropy": 4.571897840499878, "epoch": 7.277300613496933, "grad_norm": 0.59375, "learning_rate": 0.00014046531905576265, "loss": 4.0647, "mean_token_accuracy": 0.3074125051498413, "num_tokens": 144638789.0, "step": 8900 }, { "entropy": 4.596830034255982, "epoch": 7.281390593047035, "grad_norm": 0.59375, "learning_rate": 0.00014021296430013385, "loss": 4.0891, "mean_token_accuracy": 0.3050909131765366, "num_tokens": 144720486.0, "step": 8905 }, { "entropy": 4.607835721969605, "epoch": 7.285480572597137, "grad_norm": 0.58984375, "learning_rate": 0.00013996087372581637, "loss": 4.1298, "mean_token_accuracy": 0.30170731246471405, "num_tokens": 144801605.0, "step": 8910 }, { "entropy": 4.653624248504639, "epoch": 7.289570552147239, "grad_norm": 0.609375, "learning_rate": 0.00013970904782690526, "loss": 4.1679, "mean_token_accuracy": 0.29894726574420927, "num_tokens": 144883119.0, "step": 8915 }, { "entropy": 4.615035724639893, "epoch": 7.293660531697341, "grad_norm": 0.58984375, "learning_rate": 0.000139457487096977, "loss": 4.0988, "mean_token_accuracy": 0.30412229597568513, "num_tokens": 144964771.0, "step": 8920 }, { "entropy": 4.614809656143189, "epoch": 7.297750511247444, "grad_norm": 0.5859375, "learning_rate": 0.00013920619202908798, "loss": 4.1294, "mean_token_accuracy": 0.2976973295211792, "num_tokens": 145046073.0, "step": 8925 }, { "entropy": 4.592530536651611, "epoch": 7.301840490797546, "grad_norm": 0.59765625, "learning_rate": 0.00013895516311577456, "loss": 4.1042, "mean_token_accuracy": 0.30382744669914247, "num_tokens": 145127224.0, "step": 8930 }, { "entropy": 4.572969198226929, "epoch": 7.305930470347648, "grad_norm": 0.6015625, "learning_rate": 0.00013870440084905064, "loss": 4.0661, "mean_token_accuracy": 0.3073374181985855, "num_tokens": 145207823.0, "step": 8935 }, { "entropy": 4.608275890350342, "epoch": 7.310020449897751, "grad_norm": 0.59765625, "learning_rate": 0.00013845390572040823, "loss": 4.116, "mean_token_accuracy": 0.30430246889591217, "num_tokens": 145288784.0, "step": 8940 }, { "entropy": 4.594882822036743, "epoch": 7.314110429447853, "grad_norm": 0.60546875, "learning_rate": 0.00013820367822081503, "loss": 4.0795, "mean_token_accuracy": 0.3057946264743805, "num_tokens": 145370065.0, "step": 8945 }, { "entropy": 4.591764068603515, "epoch": 7.318200408997955, "grad_norm": 0.5859375, "learning_rate": 0.00013795371884071486, "loss": 4.0984, "mean_token_accuracy": 0.3020516008138657, "num_tokens": 145451240.0, "step": 8950 }, { "entropy": 4.62282919883728, "epoch": 7.3222903885480575, "grad_norm": 0.58203125, "learning_rate": 0.00013770402807002566, "loss": 4.1202, "mean_token_accuracy": 0.30376642346382143, "num_tokens": 145532933.0, "step": 8955 }, { "entropy": 4.602484846115113, "epoch": 7.326380368098159, "grad_norm": 0.58984375, "learning_rate": 0.0001374546063981389, "loss": 4.1158, "mean_token_accuracy": 0.30150844156742096, "num_tokens": 145613969.0, "step": 8960 }, { "entropy": 4.619353437423706, "epoch": 7.330470347648262, "grad_norm": 0.61328125, "learning_rate": 0.00013720545431391879, "loss": 4.1353, "mean_token_accuracy": 0.3026641994714737, "num_tokens": 145695519.0, "step": 8965 }, { "entropy": 4.615701723098755, "epoch": 7.334560327198364, "grad_norm": 0.6015625, "learning_rate": 0.00013695657230570097, "loss": 4.1327, "mean_token_accuracy": 0.2991668492555618, "num_tokens": 145777032.0, "step": 8970 }, { "entropy": 4.616499233245849, "epoch": 7.338650306748466, "grad_norm": 0.59375, "learning_rate": 0.00013670796086129184, "loss": 4.1042, "mean_token_accuracy": 0.30256164968013766, "num_tokens": 145858266.0, "step": 8975 }, { "entropy": 4.580845785140991, "epoch": 7.342740286298569, "grad_norm": 0.59765625, "learning_rate": 0.0001364596204679674, "loss": 4.0964, "mean_token_accuracy": 0.3047081887722015, "num_tokens": 145939864.0, "step": 8980 }, { "entropy": 4.6191661834716795, "epoch": 7.346830265848671, "grad_norm": 0.6015625, "learning_rate": 0.00013621155161247273, "loss": 4.1166, "mean_token_accuracy": 0.30437380373477935, "num_tokens": 146020927.0, "step": 8985 }, { "entropy": 4.669764375686645, "epoch": 7.350920245398773, "grad_norm": 0.6015625, "learning_rate": 0.00013596375478102011, "loss": 4.1813, "mean_token_accuracy": 0.29761673510074615, "num_tokens": 146102241.0, "step": 8990 }, { "entropy": 4.593724584579467, "epoch": 7.355010224948876, "grad_norm": 0.59765625, "learning_rate": 0.00013571623045928916, "loss": 4.1009, "mean_token_accuracy": 0.3050650000572205, "num_tokens": 146183820.0, "step": 8995 }, { "entropy": 4.597909259796142, "epoch": 7.3591002044989775, "grad_norm": 0.6015625, "learning_rate": 0.00013546897913242512, "loss": 4.1044, "mean_token_accuracy": 0.302783265709877, "num_tokens": 146265106.0, "step": 9000 }, { "epoch": 7.3591002044989775, "eval_entropy": 4.4404738722620785, "eval_loss": 4.585560321807861, "eval_mean_token_accuracy": 0.27040296522576834, "eval_num_tokens": 146265106.0, "eval_runtime": 68.3282, "eval_samples_per_second": 60.546, "eval_steps_per_second": 7.581, "step": 9000 } ], "logging_steps": 5, "max_steps": 12220, "num_input_tokens_seen": 0, "num_train_epochs": 10, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 7.4644349135616e+16, "train_batch_size": 16, "trial_name": null, "trial_params": null }