{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 0.8179959100204499, "eval_steps": 500, "global_step": 1000, "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 } ], "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": 8300572171776000.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }