{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 2.862167689161554, "eval_steps": 500, "global_step": 3500, "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 } ], "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": 2.9034154620288e+16, "train_batch_size": 16, "trial_name": null, "trial_params": null }