| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.2, |
| "eval_steps": 100, |
| "global_step": 2000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.1014384485781192, |
| "epoch": 0.016, |
| "grad_norm": 2.5625, |
| "learning_rate": 3.0000000000000004e-07, |
| "loss": 1.1876, |
| "mean_token_accuracy": 0.7093198902904987, |
| "num_tokens": 195510.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.1004723750054837, |
| "epoch": 0.032, |
| "grad_norm": 2.40625, |
| "learning_rate": 6.333333333333334e-07, |
| "loss": 1.1865, |
| "mean_token_accuracy": 0.7083669044077396, |
| "num_tokens": 391008.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.1084382019937038, |
| "epoch": 0.048, |
| "grad_norm": 2.640625, |
| "learning_rate": 9.666666666666668e-07, |
| "loss": 1.2086, |
| "mean_token_accuracy": 0.7065350603312254, |
| "num_tokens": 590136.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.107455987483263, |
| "epoch": 0.064, |
| "grad_norm": 2.203125, |
| "learning_rate": 1.3e-06, |
| "loss": 1.1902, |
| "mean_token_accuracy": 0.7083055440336465, |
| "num_tokens": 791432.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.1283973854035139, |
| "epoch": 0.08, |
| "grad_norm": 2.5625, |
| "learning_rate": 1.6333333333333335e-06, |
| "loss": 1.1938, |
| "mean_token_accuracy": 0.7073600720614195, |
| "num_tokens": 990342.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.1554226335138083, |
| "epoch": 0.096, |
| "grad_norm": 2.546875, |
| "learning_rate": 1.9666666666666668e-06, |
| "loss": 1.2297, |
| "mean_token_accuracy": 0.7054298754781485, |
| "num_tokens": 1182339.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.1388252694159746, |
| "epoch": 0.112, |
| "grad_norm": 2.40625, |
| "learning_rate": 2.3000000000000004e-06, |
| "loss": 1.1945, |
| "mean_token_accuracy": 0.70603371784091, |
| "num_tokens": 1386193.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.138149081543088, |
| "epoch": 0.128, |
| "grad_norm": 2.5, |
| "learning_rate": 2.6333333333333332e-06, |
| "loss": 1.1892, |
| "mean_token_accuracy": 0.7073806554079056, |
| "num_tokens": 1583261.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.154042448103428, |
| "epoch": 0.144, |
| "grad_norm": 2.296875, |
| "learning_rate": 2.9666666666666673e-06, |
| "loss": 1.2091, |
| "mean_token_accuracy": 0.7068144842982292, |
| "num_tokens": 1774826.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.155431379005313, |
| "epoch": 0.16, |
| "grad_norm": 2.984375, |
| "learning_rate": 3.3000000000000006e-06, |
| "loss": 1.2129, |
| "mean_token_accuracy": 0.7055633965879679, |
| "num_tokens": 1971310.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 1.1157788362503052, |
| "eval_biology_loss": 1.3415579795837402, |
| "eval_biology_mean_token_accuracy": 0.6915143814086914, |
| "eval_biology_num_tokens": 1971310.0, |
| "eval_biology_runtime": 27.5411, |
| "eval_biology_samples_per_second": 18.155, |
| "eval_biology_steps_per_second": 4.539, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 0.9838901553153991, |
| "eval_chemistry_loss": 1.298728585243225, |
| "eval_chemistry_mean_token_accuracy": 0.7006946277618408, |
| "eval_chemistry_num_tokens": 1971310.0, |
| "eval_chemistry_runtime": 34.8186, |
| "eval_chemistry_samples_per_second": 14.36, |
| "eval_chemistry_steps_per_second": 3.59, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 1.1278482675552368, |
| "eval_math_loss": 1.8617346286773682, |
| "eval_math_mean_token_accuracy": 0.6164904050827026, |
| "eval_math_num_tokens": 1971310.0, |
| "eval_math_runtime": 35.8517, |
| "eval_math_samples_per_second": 13.946, |
| "eval_math_steps_per_second": 3.487, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 1.0754256196022034, |
| "eval_physics_loss": 1.4391310214996338, |
| "eval_physics_mean_token_accuracy": 0.679765085220337, |
| "eval_physics_num_tokens": 1971310.0, |
| "eval_physics_runtime": 41.5159, |
| "eval_physics_samples_per_second": 12.044, |
| "eval_physics_steps_per_second": 3.011, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.1395905867218972, |
| "epoch": 0.176, |
| "grad_norm": 2.265625, |
| "learning_rate": 3.633333333333334e-06, |
| "loss": 1.1892, |
| "mean_token_accuracy": 0.708667328581214, |
| "num_tokens": 2169453.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.131948610395193, |
| "epoch": 0.192, |
| "grad_norm": 2.484375, |
| "learning_rate": 3.966666666666667e-06, |
| "loss": 1.1924, |
| "mean_token_accuracy": 0.70796740911901, |
| "num_tokens": 2366844.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.1465655714273453, |
| "epoch": 0.208, |
| "grad_norm": 2.546875, |
| "learning_rate": 4.3e-06, |
| "loss": 1.2003, |
| "mean_token_accuracy": 0.7059498965740204, |
| "num_tokens": 2559708.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.1307410310953856, |
| "epoch": 0.224, |
| "grad_norm": 2.546875, |
| "learning_rate": 4.633333333333334e-06, |
| "loss": 1.1966, |
| "mean_token_accuracy": 0.7108498476445675, |
| "num_tokens": 2756533.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.1295269019901752, |
| "epoch": 0.24, |
| "grad_norm": 2.578125, |
| "learning_rate": 4.966666666666667e-06, |
| "loss": 1.1845, |
| "mean_token_accuracy": 0.7088116876780987, |
| "num_tokens": 2948600.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.1384097158908844, |
| "epoch": 0.256, |
| "grad_norm": 2.3125, |
| "learning_rate": 5.300000000000001e-06, |
| "loss": 1.1861, |
| "mean_token_accuracy": 0.709713039174676, |
| "num_tokens": 3141473.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.1169631101191044, |
| "epoch": 0.272, |
| "grad_norm": 2.546875, |
| "learning_rate": 5.633333333333334e-06, |
| "loss": 1.1727, |
| "mean_token_accuracy": 0.7096305072307587, |
| "num_tokens": 3337285.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.1154247954487801, |
| "epoch": 0.288, |
| "grad_norm": 2.46875, |
| "learning_rate": 5.966666666666667e-06, |
| "loss": 1.1645, |
| "mean_token_accuracy": 0.7108631394803524, |
| "num_tokens": 3541221.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.1156449727714062, |
| "epoch": 0.304, |
| "grad_norm": 2.453125, |
| "learning_rate": 6.300000000000001e-06, |
| "loss": 1.1768, |
| "mean_token_accuracy": 0.7135872103273868, |
| "num_tokens": 3734952.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.137134049460292, |
| "epoch": 0.32, |
| "grad_norm": 2.515625, |
| "learning_rate": 6.633333333333334e-06, |
| "loss": 1.1918, |
| "mean_token_accuracy": 0.7095682512968778, |
| "num_tokens": 3922209.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 1.103229732990265, |
| "eval_biology_loss": 1.3444453477859497, |
| "eval_biology_mean_token_accuracy": 0.6913131108283996, |
| "eval_biology_num_tokens": 3922209.0, |
| "eval_biology_runtime": 27.5062, |
| "eval_biology_samples_per_second": 18.178, |
| "eval_biology_steps_per_second": 4.544, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 0.9783513498306274, |
| "eval_chemistry_loss": 1.3024827241897583, |
| "eval_chemistry_mean_token_accuracy": 0.7001587204933166, |
| "eval_chemistry_num_tokens": 3922209.0, |
| "eval_chemistry_runtime": 34.7679, |
| "eval_chemistry_samples_per_second": 14.381, |
| "eval_chemistry_steps_per_second": 3.595, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 1.1308464035987853, |
| "eval_math_loss": 1.862917423248291, |
| "eval_math_mean_token_accuracy": 0.6154395718574523, |
| "eval_math_num_tokens": 3922209.0, |
| "eval_math_runtime": 35.8096, |
| "eval_math_samples_per_second": 13.963, |
| "eval_math_steps_per_second": 3.491, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 1.0741513648033143, |
| "eval_physics_loss": 1.4400166273117065, |
| "eval_physics_mean_token_accuracy": 0.6793102626800537, |
| "eval_physics_num_tokens": 3922209.0, |
| "eval_physics_runtime": 41.5075, |
| "eval_physics_samples_per_second": 12.046, |
| "eval_physics_steps_per_second": 3.012, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.0870313856750726, |
| "epoch": 0.336, |
| "grad_norm": 2.40625, |
| "learning_rate": 6.966666666666667e-06, |
| "loss": 1.153, |
| "mean_token_accuracy": 0.717507753148675, |
| "num_tokens": 4115941.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.1046641763299703, |
| "epoch": 0.352, |
| "grad_norm": 2.40625, |
| "learning_rate": 7.3e-06, |
| "loss": 1.1538, |
| "mean_token_accuracy": 0.7155704971402883, |
| "num_tokens": 4308755.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 1.1052337765693665, |
| "epoch": 0.368, |
| "grad_norm": 2.4375, |
| "learning_rate": 7.633333333333334e-06, |
| "loss": 1.1641, |
| "mean_token_accuracy": 0.7153849575668574, |
| "num_tokens": 4505910.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.1033248636871575, |
| "epoch": 0.384, |
| "grad_norm": 2.53125, |
| "learning_rate": 7.966666666666668e-06, |
| "loss": 1.1557, |
| "mean_token_accuracy": 0.7156476765871048, |
| "num_tokens": 4696101.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.1119797427207232, |
| "epoch": 0.4, |
| "grad_norm": 2.40625, |
| "learning_rate": 8.3e-06, |
| "loss": 1.172, |
| "mean_token_accuracy": 0.7119910713285208, |
| "num_tokens": 4889445.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 1.1252103634178638, |
| "epoch": 0.416, |
| "grad_norm": 2.296875, |
| "learning_rate": 8.633333333333334e-06, |
| "loss": 1.1665, |
| "mean_token_accuracy": 0.7105265349149704, |
| "num_tokens": 5087624.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 1.1076968360692263, |
| "epoch": 0.432, |
| "grad_norm": 2.625, |
| "learning_rate": 8.966666666666667e-06, |
| "loss": 1.1673, |
| "mean_token_accuracy": 0.7133604180067777, |
| "num_tokens": 5273794.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 1.1073474381119013, |
| "epoch": 0.448, |
| "grad_norm": 2.6875, |
| "learning_rate": 9.3e-06, |
| "loss": 1.1731, |
| "mean_token_accuracy": 0.7120872918516398, |
| "num_tokens": 5463288.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 1.08086390607059, |
| "epoch": 0.464, |
| "grad_norm": 2.546875, |
| "learning_rate": 9.633333333333335e-06, |
| "loss": 1.134, |
| "mean_token_accuracy": 0.7202123641967774, |
| "num_tokens": 5656827.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 1.0910722456872464, |
| "epoch": 0.48, |
| "grad_norm": 2.359375, |
| "learning_rate": 9.966666666666667e-06, |
| "loss": 1.153, |
| "mean_token_accuracy": 0.7153508011251688, |
| "num_tokens": 5853311.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 1.0870987777709962, |
| "eval_biology_loss": 1.3525676727294922, |
| "eval_biology_mean_token_accuracy": 0.6901911664009094, |
| "eval_biology_num_tokens": 5853311.0, |
| "eval_biology_runtime": 27.5183, |
| "eval_biology_samples_per_second": 18.17, |
| "eval_biology_steps_per_second": 4.542, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 0.9616376605033875, |
| "eval_chemistry_loss": 1.321352243423462, |
| "eval_chemistry_mean_token_accuracy": 0.6977167620658874, |
| "eval_chemistry_num_tokens": 5853311.0, |
| "eval_chemistry_runtime": 34.7786, |
| "eval_chemistry_samples_per_second": 14.377, |
| "eval_chemistry_steps_per_second": 3.594, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 1.115381636619568, |
| "eval_math_loss": 1.895226001739502, |
| "eval_math_mean_token_accuracy": 0.6132121315002441, |
| "eval_math_num_tokens": 5853311.0, |
| "eval_math_runtime": 35.8243, |
| "eval_math_samples_per_second": 13.957, |
| "eval_math_steps_per_second": 3.489, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 1.0583682708740234, |
| "eval_physics_loss": 1.470627784729004, |
| "eval_physics_mean_token_accuracy": 0.675459557056427, |
| "eval_physics_num_tokens": 5853311.0, |
| "eval_physics_runtime": 41.5097, |
| "eval_physics_samples_per_second": 12.045, |
| "eval_physics_steps_per_second": 3.011, |
| "step": 300 |
| }, |
| { |
| "entropy": 1.034109428524971, |
| "epoch": 0.496, |
| "grad_norm": 2.515625, |
| "learning_rate": 1.0300000000000001e-05, |
| "loss": 1.0884, |
| "mean_token_accuracy": 0.7270658195018769, |
| "num_tokens": 6049788.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 1.0862518075853587, |
| "epoch": 0.512, |
| "grad_norm": 2.296875, |
| "learning_rate": 1.0633333333333334e-05, |
| "loss": 1.143, |
| "mean_token_accuracy": 0.7176959495991468, |
| "num_tokens": 6243825.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 1.0936558537185193, |
| "epoch": 0.528, |
| "grad_norm": 2.421875, |
| "learning_rate": 1.0966666666666668e-05, |
| "loss": 1.1506, |
| "mean_token_accuracy": 0.7166993264108896, |
| "num_tokens": 6434188.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 1.0839752975851298, |
| "epoch": 0.544, |
| "grad_norm": 2.25, |
| "learning_rate": 1.13e-05, |
| "loss": 1.1531, |
| "mean_token_accuracy": 0.7193178791552782, |
| "num_tokens": 6629832.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 1.083231647312641, |
| "epoch": 0.56, |
| "grad_norm": 2.375, |
| "learning_rate": 1.1633333333333334e-05, |
| "loss": 1.1463, |
| "mean_token_accuracy": 0.7157608140259981, |
| "num_tokens": 6824668.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 1.0830022376030684, |
| "epoch": 0.576, |
| "grad_norm": 2.375, |
| "learning_rate": 1.1966666666666668e-05, |
| "loss": 1.1328, |
| "mean_token_accuracy": 0.7173026569187642, |
| "num_tokens": 7025923.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 1.0897728819400072, |
| "epoch": 0.592, |
| "grad_norm": 2.609375, |
| "learning_rate": 1.23e-05, |
| "loss": 1.1455, |
| "mean_token_accuracy": 0.7154262211173773, |
| "num_tokens": 7218224.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 1.071556494385004, |
| "epoch": 0.608, |
| "grad_norm": 2.5625, |
| "learning_rate": 1.2633333333333335e-05, |
| "loss": 1.1397, |
| "mean_token_accuracy": 0.719798355922103, |
| "num_tokens": 7421216.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 1.099830712750554, |
| "epoch": 0.624, |
| "grad_norm": 2.46875, |
| "learning_rate": 1.2966666666666667e-05, |
| "loss": 1.1435, |
| "mean_token_accuracy": 0.7161061268299818, |
| "num_tokens": 7617258.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 1.1015911962836982, |
| "epoch": 0.64, |
| "grad_norm": 2.5625, |
| "learning_rate": 1.3300000000000001e-05, |
| "loss": 1.1593, |
| "mean_token_accuracy": 0.7135881155729293, |
| "num_tokens": 7806173.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 1.0770543112754822, |
| "eval_biology_loss": 1.3552216291427612, |
| "eval_biology_mean_token_accuracy": 0.6905452556610108, |
| "eval_biology_num_tokens": 7806173.0, |
| "eval_biology_runtime": 27.5408, |
| "eval_biology_samples_per_second": 18.155, |
| "eval_biology_steps_per_second": 4.539, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 0.9578963780403137, |
| "eval_chemistry_loss": 1.3227319717407227, |
| "eval_chemistry_mean_token_accuracy": 0.6978275780677795, |
| "eval_chemistry_num_tokens": 7806173.0, |
| "eval_chemistry_runtime": 34.7767, |
| "eval_chemistry_samples_per_second": 14.377, |
| "eval_chemistry_steps_per_second": 3.594, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 1.1102556204795837, |
| "eval_math_loss": 1.9120879173278809, |
| "eval_math_mean_token_accuracy": 0.6123861393928528, |
| "eval_math_num_tokens": 7806173.0, |
| "eval_math_runtime": 35.829, |
| "eval_math_samples_per_second": 13.955, |
| "eval_math_steps_per_second": 3.489, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 1.0522716245651245, |
| "eval_physics_loss": 1.470394253730774, |
| "eval_physics_mean_token_accuracy": 0.6759520478248596, |
| "eval_physics_num_tokens": 7806173.0, |
| "eval_physics_runtime": 41.5378, |
| "eval_physics_samples_per_second": 12.037, |
| "eval_physics_steps_per_second": 3.009, |
| "step": 400 |
| }, |
| { |
| "entropy": 1.0875599801540374, |
| "epoch": 0.656, |
| "grad_norm": 2.515625, |
| "learning_rate": 1.3633333333333334e-05, |
| "loss": 1.1542, |
| "mean_token_accuracy": 0.7155895106494427, |
| "num_tokens": 8000514.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 1.107854114100337, |
| "epoch": 0.672, |
| "grad_norm": 2.65625, |
| "learning_rate": 1.3966666666666668e-05, |
| "loss": 1.1627, |
| "mean_token_accuracy": 0.7143586546182632, |
| "num_tokens": 8187884.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 1.0949572917073964, |
| "epoch": 0.688, |
| "grad_norm": 2.3125, |
| "learning_rate": 1.43e-05, |
| "loss": 1.1528, |
| "mean_token_accuracy": 0.7152665574103594, |
| "num_tokens": 8390643.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 1.0888024061918258, |
| "epoch": 0.704, |
| "grad_norm": 2.5, |
| "learning_rate": 1.4633333333333334e-05, |
| "loss": 1.1436, |
| "mean_token_accuracy": 0.7173040930181742, |
| "num_tokens": 8583266.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 1.0663380671292544, |
| "epoch": 0.72, |
| "grad_norm": 2.671875, |
| "learning_rate": 1.4966666666666667e-05, |
| "loss": 1.1062, |
| "mean_token_accuracy": 0.7224690090864897, |
| "num_tokens": 8786233.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 1.070183601230383, |
| "epoch": 0.736, |
| "grad_norm": 2.28125, |
| "learning_rate": 1.5300000000000003e-05, |
| "loss": 1.1219, |
| "mean_token_accuracy": 0.7191323135048151, |
| "num_tokens": 8982898.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 1.0922042228281499, |
| "epoch": 0.752, |
| "grad_norm": 2.40625, |
| "learning_rate": 1.5633333333333335e-05, |
| "loss": 1.1567, |
| "mean_token_accuracy": 0.7144051775336265, |
| "num_tokens": 9174361.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 1.071123208478093, |
| "epoch": 0.768, |
| "grad_norm": 2.578125, |
| "learning_rate": 1.5966666666666667e-05, |
| "loss": 1.127, |
| "mean_token_accuracy": 0.7194326717406512, |
| "num_tokens": 9373430.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 1.0949943471699952, |
| "epoch": 0.784, |
| "grad_norm": 2.640625, |
| "learning_rate": 1.63e-05, |
| "loss": 1.142, |
| "mean_token_accuracy": 0.716059061512351, |
| "num_tokens": 9570409.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 1.130943875014782, |
| "epoch": 0.8, |
| "grad_norm": 2.734375, |
| "learning_rate": 1.6633333333333336e-05, |
| "loss": 1.1826, |
| "mean_token_accuracy": 0.7076376970857382, |
| "num_tokens": 9766360.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 1.0730011610984802, |
| "eval_biology_loss": 1.3595266342163086, |
| "eval_biology_mean_token_accuracy": 0.6896284861564637, |
| "eval_biology_num_tokens": 9766360.0, |
| "eval_biology_runtime": 27.5076, |
| "eval_biology_samples_per_second": 18.177, |
| "eval_biology_steps_per_second": 4.544, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 0.9538066201210021, |
| "eval_chemistry_loss": 1.3289003372192383, |
| "eval_chemistry_mean_token_accuracy": 0.696358925819397, |
| "eval_chemistry_num_tokens": 9766360.0, |
| "eval_chemistry_runtime": 34.5073, |
| "eval_chemistry_samples_per_second": 14.49, |
| "eval_chemistry_steps_per_second": 3.622, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 1.143517897129059, |
| "eval_math_loss": 1.90286386013031, |
| "eval_math_mean_token_accuracy": 0.6096776604652405, |
| "eval_math_num_tokens": 9766360.0, |
| "eval_math_runtime": 35.6555, |
| "eval_math_samples_per_second": 14.023, |
| "eval_math_steps_per_second": 3.506, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 1.061961835861206, |
| "eval_physics_loss": 1.4838680028915405, |
| "eval_physics_mean_token_accuracy": 0.6727198429107666, |
| "eval_physics_num_tokens": 9766360.0, |
| "eval_physics_runtime": 41.4583, |
| "eval_physics_samples_per_second": 12.06, |
| "eval_physics_steps_per_second": 3.015, |
| "step": 500 |
| }, |
| { |
| "entropy": 1.0811156008392573, |
| "epoch": 0.816, |
| "grad_norm": 2.765625, |
| "learning_rate": 1.6966666666666668e-05, |
| "loss": 1.1388, |
| "mean_token_accuracy": 0.7171396769583225, |
| "num_tokens": 9963834.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 1.0769522741436959, |
| "epoch": 0.832, |
| "grad_norm": 2.625, |
| "learning_rate": 1.73e-05, |
| "loss": 1.1431, |
| "mean_token_accuracy": 0.7170515075325966, |
| "num_tokens": 10160950.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 1.0829918976873159, |
| "epoch": 0.848, |
| "grad_norm": 2.9375, |
| "learning_rate": 1.7633333333333336e-05, |
| "loss": 1.135, |
| "mean_token_accuracy": 0.7176959712058306, |
| "num_tokens": 10363131.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 1.1059647228568792, |
| "epoch": 0.864, |
| "grad_norm": 2.953125, |
| "learning_rate": 1.796666666666667e-05, |
| "loss": 1.164, |
| "mean_token_accuracy": 0.7122997116297484, |
| "num_tokens": 10558003.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 1.0871139224618673, |
| "epoch": 0.88, |
| "grad_norm": 2.515625, |
| "learning_rate": 1.83e-05, |
| "loss": 1.1461, |
| "mean_token_accuracy": 0.7161415867507458, |
| "num_tokens": 10754470.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 1.1228669207543134, |
| "epoch": 0.896, |
| "grad_norm": 2.859375, |
| "learning_rate": 1.8633333333333333e-05, |
| "loss": 1.1752, |
| "mean_token_accuracy": 0.7109104376286268, |
| "num_tokens": 10948943.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 1.0997110158205032, |
| "epoch": 0.912, |
| "grad_norm": 2.453125, |
| "learning_rate": 1.896666666666667e-05, |
| "loss": 1.1551, |
| "mean_token_accuracy": 0.7136997357010841, |
| "num_tokens": 11142613.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 1.1056707065552474, |
| "epoch": 0.928, |
| "grad_norm": 2.390625, |
| "learning_rate": 1.93e-05, |
| "loss": 1.1579, |
| "mean_token_accuracy": 0.7122712273150682, |
| "num_tokens": 11336779.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 1.0623173102736474, |
| "epoch": 0.944, |
| "grad_norm": 2.359375, |
| "learning_rate": 1.9633333333333334e-05, |
| "loss": 1.1309, |
| "mean_token_accuracy": 0.720225491374731, |
| "num_tokens": 11536381.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 1.0885680321604014, |
| "epoch": 0.96, |
| "grad_norm": 2.5, |
| "learning_rate": 1.9966666666666666e-05, |
| "loss": 1.1453, |
| "mean_token_accuracy": 0.715793015062809, |
| "num_tokens": 11735527.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 1.0950671820640563, |
| "eval_biology_loss": 1.3571794033050537, |
| "eval_biology_mean_token_accuracy": 0.6889836673736572, |
| "eval_biology_num_tokens": 11735527.0, |
| "eval_biology_runtime": 27.5045, |
| "eval_biology_samples_per_second": 18.179, |
| "eval_biology_steps_per_second": 4.545, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_chemistry_entropy": 0.9710470175743103, |
| "eval_chemistry_loss": 1.3357082605361938, |
| "eval_chemistry_mean_token_accuracy": 0.694826488494873, |
| "eval_chemistry_num_tokens": 11735527.0, |
| "eval_chemistry_runtime": 34.7827, |
| "eval_chemistry_samples_per_second": 14.375, |
| "eval_chemistry_steps_per_second": 3.594, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_math_entropy": 1.153974467277527, |
| "eval_math_loss": 1.9200637340545654, |
| "eval_math_mean_token_accuracy": 0.6068759155273438, |
| "eval_math_num_tokens": 11735527.0, |
| "eval_math_runtime": 35.8443, |
| "eval_math_samples_per_second": 13.949, |
| "eval_math_steps_per_second": 3.487, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_physics_entropy": 1.0831837186813353, |
| "eval_physics_loss": 1.4995465278625488, |
| "eval_physics_mean_token_accuracy": 0.6700002565383911, |
| "eval_physics_num_tokens": 11735527.0, |
| "eval_physics_runtime": 41.5429, |
| "eval_physics_samples_per_second": 12.036, |
| "eval_physics_steps_per_second": 3.009, |
| "step": 600 |
| }, |
| { |
| "entropy": 1.0806872766464948, |
| "epoch": 0.976, |
| "grad_norm": 2.515625, |
| "learning_rate": 1.9966666666666666e-05, |
| "loss": 1.1385, |
| "mean_token_accuracy": 0.7172170367091895, |
| "num_tokens": 11930935.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 1.1068653509020805, |
| "epoch": 0.992, |
| "grad_norm": 2.421875, |
| "learning_rate": 1.992962962962963e-05, |
| "loss": 1.1719, |
| "mean_token_accuracy": 0.7108269903808833, |
| "num_tokens": 12129585.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 1.098480136319995, |
| "epoch": 1.008, |
| "grad_norm": 2.359375, |
| "learning_rate": 1.9892592592592593e-05, |
| "loss": 1.1499, |
| "mean_token_accuracy": 0.7172095529735089, |
| "num_tokens": 12325847.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 1.046534773334861, |
| "epoch": 1.024, |
| "grad_norm": 2.53125, |
| "learning_rate": 1.9855555555555558e-05, |
| "loss": 1.1007, |
| "mean_token_accuracy": 0.7235685538500547, |
| "num_tokens": 12530820.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 1.0832492623478174, |
| "epoch": 1.04, |
| "grad_norm": 2.546875, |
| "learning_rate": 1.981851851851852e-05, |
| "loss": 1.1431, |
| "mean_token_accuracy": 0.7168577324599028, |
| "num_tokens": 12725384.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 1.0526921059936285, |
| "epoch": 1.056, |
| "grad_norm": 2.515625, |
| "learning_rate": 1.9781481481481484e-05, |
| "loss": 1.1021, |
| "mean_token_accuracy": 0.7233843062072992, |
| "num_tokens": 12924592.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 1.074445153772831, |
| "epoch": 1.072, |
| "grad_norm": 2.640625, |
| "learning_rate": 1.9744444444444446e-05, |
| "loss": 1.1325, |
| "mean_token_accuracy": 0.7200184687972069, |
| "num_tokens": 13112848.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 1.0619338382035495, |
| "epoch": 1.088, |
| "grad_norm": 2.734375, |
| "learning_rate": 1.9707407407407407e-05, |
| "loss": 1.1287, |
| "mean_token_accuracy": 0.7215401880443096, |
| "num_tokens": 13305725.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 1.0723260939121246, |
| "epoch": 1.104, |
| "grad_norm": 2.578125, |
| "learning_rate": 1.9670370370370372e-05, |
| "loss": 1.1277, |
| "mean_token_accuracy": 0.7210916347801686, |
| "num_tokens": 13498685.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 1.0689454402774572, |
| "epoch": 1.12, |
| "grad_norm": 2.875, |
| "learning_rate": 1.9633333333333334e-05, |
| "loss": 1.1341, |
| "mean_token_accuracy": 0.7211067486554384, |
| "num_tokens": 13682146.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_biology_entropy": 1.052022078514099, |
| "eval_biology_loss": 1.3665521144866943, |
| "eval_biology_mean_token_accuracy": 0.6888244609832763, |
| "eval_biology_num_tokens": 13682146.0, |
| "eval_biology_runtime": 27.5122, |
| "eval_biology_samples_per_second": 18.174, |
| "eval_biology_steps_per_second": 4.543, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_chemistry_entropy": 0.9264287524223328, |
| "eval_chemistry_loss": 1.360307216644287, |
| "eval_chemistry_mean_token_accuracy": 0.6919433879852295, |
| "eval_chemistry_num_tokens": 13682146.0, |
| "eval_chemistry_runtime": 34.7603, |
| "eval_chemistry_samples_per_second": 14.384, |
| "eval_chemistry_steps_per_second": 3.596, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_math_entropy": 1.0819805884361267, |
| "eval_math_loss": 1.9619777202606201, |
| "eval_math_mean_token_accuracy": 0.603320020198822, |
| "eval_math_num_tokens": 13682146.0, |
| "eval_math_runtime": 35.8109, |
| "eval_math_samples_per_second": 13.962, |
| "eval_math_steps_per_second": 3.491, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_physics_entropy": 1.0298279185295105, |
| "eval_physics_loss": 1.5183902978897095, |
| "eval_physics_mean_token_accuracy": 0.6680226521492004, |
| "eval_physics_num_tokens": 13682146.0, |
| "eval_physics_runtime": 41.5476, |
| "eval_physics_samples_per_second": 12.034, |
| "eval_physics_steps_per_second": 3.009, |
| "step": 700 |
| }, |
| { |
| "entropy": 1.0617417309433221, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 2.671875, |
| "learning_rate": 1.95962962962963e-05, |
| "loss": 1.1242, |
| "mean_token_accuracy": 0.7208964262157679, |
| "num_tokens": 13876656.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 1.0505924891680478, |
| "epoch": 1.152, |
| "grad_norm": 2.484375, |
| "learning_rate": 1.955925925925926e-05, |
| "loss": 1.0977, |
| "mean_token_accuracy": 0.7227281857281923, |
| "num_tokens": 14086104.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 1.0598271984606982, |
| "epoch": 1.168, |
| "grad_norm": 2.3125, |
| "learning_rate": 1.9522222222222225e-05, |
| "loss": 1.1215, |
| "mean_token_accuracy": 0.7210871536284685, |
| "num_tokens": 14274520.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 1.0640012204647065, |
| "epoch": 1.184, |
| "grad_norm": 2.609375, |
| "learning_rate": 1.9485185185185187e-05, |
| "loss": 1.1269, |
| "mean_token_accuracy": 0.7209575697779655, |
| "num_tokens": 14473623.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 1.061717476695776, |
| "epoch": 1.2, |
| "grad_norm": 2.625, |
| "learning_rate": 1.944814814814815e-05, |
| "loss": 1.1264, |
| "mean_token_accuracy": 0.7198022853583097, |
| "num_tokens": 14661238.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 1.036321223527193, |
| "epoch": 1.216, |
| "grad_norm": 2.5, |
| "learning_rate": 1.941111111111111e-05, |
| "loss": 1.0904, |
| "mean_token_accuracy": 0.7260100528597832, |
| "num_tokens": 14865915.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 1.058328016847372, |
| "epoch": 1.232, |
| "grad_norm": 2.390625, |
| "learning_rate": 1.9374074074074075e-05, |
| "loss": 1.1201, |
| "mean_token_accuracy": 0.7228422287851572, |
| "num_tokens": 15055682.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 1.0587275315076112, |
| "epoch": 1.248, |
| "grad_norm": 2.484375, |
| "learning_rate": 1.933703703703704e-05, |
| "loss": 1.128, |
| "mean_token_accuracy": 0.7206286292523145, |
| "num_tokens": 15249576.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 1.0823485005646944, |
| "epoch": 1.264, |
| "grad_norm": 2.296875, |
| "learning_rate": 1.93e-05, |
| "loss": 1.1349, |
| "mean_token_accuracy": 0.7188280873000622, |
| "num_tokens": 15446149.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 1.107851380109787, |
| "epoch": 1.28, |
| "grad_norm": 2.59375, |
| "learning_rate": 1.9262962962962967e-05, |
| "loss": 1.1698, |
| "mean_token_accuracy": 0.7131990294903516, |
| "num_tokens": 15639317.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_biology_entropy": 1.0596792635917665, |
| "eval_biology_loss": 1.3625826835632324, |
| "eval_biology_mean_token_accuracy": 0.6896568837165833, |
| "eval_biology_num_tokens": 15639317.0, |
| "eval_biology_runtime": 27.5131, |
| "eval_biology_samples_per_second": 18.173, |
| "eval_biology_steps_per_second": 4.543, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_chemistry_entropy": 0.9432493863105774, |
| "eval_chemistry_loss": 1.3452301025390625, |
| "eval_chemistry_mean_token_accuracy": 0.6950167322158813, |
| "eval_chemistry_num_tokens": 15639317.0, |
| "eval_chemistry_runtime": 34.7817, |
| "eval_chemistry_samples_per_second": 14.375, |
| "eval_chemistry_steps_per_second": 3.594, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_math_entropy": 1.106888451576233, |
| "eval_math_loss": 1.9185564517974854, |
| "eval_math_mean_token_accuracy": 0.6096730709075928, |
| "eval_math_num_tokens": 15639317.0, |
| "eval_math_runtime": 35.7864, |
| "eval_math_samples_per_second": 13.972, |
| "eval_math_steps_per_second": 3.493, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_physics_entropy": 1.0418448238372802, |
| "eval_physics_loss": 1.507297396659851, |
| "eval_physics_mean_token_accuracy": 0.6708806357383728, |
| "eval_physics_num_tokens": 15639317.0, |
| "eval_physics_runtime": 41.5226, |
| "eval_physics_samples_per_second": 12.042, |
| "eval_physics_steps_per_second": 3.01, |
| "step": 800 |
| }, |
| { |
| "entropy": 1.085048996657133, |
| "epoch": 1.296, |
| "grad_norm": 2.5625, |
| "learning_rate": 1.9225925925925928e-05, |
| "loss": 1.1457, |
| "mean_token_accuracy": 0.7162241782993078, |
| "num_tokens": 15835805.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 1.0919453758746385, |
| "epoch": 1.312, |
| "grad_norm": 2.53125, |
| "learning_rate": 1.918888888888889e-05, |
| "loss": 1.145, |
| "mean_token_accuracy": 0.7146618235856295, |
| "num_tokens": 16028375.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 1.085938362404704, |
| "epoch": 1.328, |
| "grad_norm": 2.4375, |
| "learning_rate": 1.915185185185185e-05, |
| "loss": 1.1422, |
| "mean_token_accuracy": 0.7159410495311022, |
| "num_tokens": 16230555.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 1.056251646205783, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 2.484375, |
| "learning_rate": 1.9114814814814816e-05, |
| "loss": 1.1146, |
| "mean_token_accuracy": 0.7235266748815775, |
| "num_tokens": 16436297.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 1.088941215723753, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 2.625, |
| "learning_rate": 1.907777777777778e-05, |
| "loss": 1.1367, |
| "mean_token_accuracy": 0.7174329798668623, |
| "num_tokens": 16630563.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 1.1173061154782773, |
| "epoch": 1.376, |
| "grad_norm": 2.765625, |
| "learning_rate": 1.9040740740740743e-05, |
| "loss": 1.171, |
| "mean_token_accuracy": 0.711220956966281, |
| "num_tokens": 16822602.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 1.0744022823870183, |
| "epoch": 1.392, |
| "grad_norm": 2.359375, |
| "learning_rate": 1.9003703703703704e-05, |
| "loss": 1.1472, |
| "mean_token_accuracy": 0.7182805575430393, |
| "num_tokens": 17015750.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 1.0772326588630676, |
| "epoch": 1.408, |
| "grad_norm": 2.125, |
| "learning_rate": 1.896666666666667e-05, |
| "loss": 1.1396, |
| "mean_token_accuracy": 0.7178659070283174, |
| "num_tokens": 17207248.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 1.0965230137109756, |
| "epoch": 1.424, |
| "grad_norm": 2.625, |
| "learning_rate": 1.892962962962963e-05, |
| "loss": 1.1586, |
| "mean_token_accuracy": 0.7151195332407951, |
| "num_tokens": 17403869.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 1.09039566218853, |
| "epoch": 1.44, |
| "grad_norm": 2.890625, |
| "learning_rate": 1.8892592592592592e-05, |
| "loss": 1.1451, |
| "mean_token_accuracy": 0.7158240955322981, |
| "num_tokens": 17597428.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_biology_entropy": 1.0646971983909608, |
| "eval_biology_loss": 1.3550127744674683, |
| "eval_biology_mean_token_accuracy": 0.6910682153701783, |
| "eval_biology_num_tokens": 17597428.0, |
| "eval_biology_runtime": 27.5142, |
| "eval_biology_samples_per_second": 18.172, |
| "eval_biology_steps_per_second": 4.543, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_chemistry_entropy": 0.9313478088378906, |
| "eval_chemistry_loss": 1.3364003896713257, |
| "eval_chemistry_mean_token_accuracy": 0.6970981850624084, |
| "eval_chemistry_num_tokens": 17597428.0, |
| "eval_chemistry_runtime": 34.7667, |
| "eval_chemistry_samples_per_second": 14.382, |
| "eval_chemistry_steps_per_second": 3.595, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_math_entropy": 1.08159548664093, |
| "eval_math_loss": 1.926034927368164, |
| "eval_math_mean_token_accuracy": 0.6117868111133575, |
| "eval_math_num_tokens": 17597428.0, |
| "eval_math_runtime": 35.8124, |
| "eval_math_samples_per_second": 13.962, |
| "eval_math_steps_per_second": 3.49, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_physics_entropy": 1.0224871401786804, |
| "eval_physics_loss": 1.4874769449234009, |
| "eval_physics_mean_token_accuracy": 0.6755724368095398, |
| "eval_physics_num_tokens": 17597428.0, |
| "eval_physics_runtime": 41.5328, |
| "eval_physics_samples_per_second": 12.039, |
| "eval_physics_steps_per_second": 3.01, |
| "step": 900 |
| }, |
| { |
| "entropy": 1.0821634780615568, |
| "epoch": 1.456, |
| "grad_norm": 2.40625, |
| "learning_rate": 1.8855555555555557e-05, |
| "loss": 1.1475, |
| "mean_token_accuracy": 0.716392119973898, |
| "num_tokens": 17798083.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 1.0787286777049303, |
| "epoch": 1.472, |
| "grad_norm": 2.421875, |
| "learning_rate": 1.881851851851852e-05, |
| "loss": 1.1325, |
| "mean_token_accuracy": 0.7189726900309324, |
| "num_tokens": 17993802.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 1.0824173912405968, |
| "epoch": 1.488, |
| "grad_norm": 2.53125, |
| "learning_rate": 1.8781481481481484e-05, |
| "loss": 1.152, |
| "mean_token_accuracy": 0.7187057230621576, |
| "num_tokens": 18185570.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 1.0912920907139778, |
| "epoch": 1.504, |
| "grad_norm": 2.3125, |
| "learning_rate": 1.8744444444444445e-05, |
| "loss": 1.1543, |
| "mean_token_accuracy": 0.7162840195000172, |
| "num_tokens": 18377903.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 1.0929523311555385, |
| "epoch": 1.52, |
| "grad_norm": 2.34375, |
| "learning_rate": 1.870740740740741e-05, |
| "loss": 1.1542, |
| "mean_token_accuracy": 0.7155724234879017, |
| "num_tokens": 18579284.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 1.1010100591927767, |
| "epoch": 1.536, |
| "grad_norm": 2.5, |
| "learning_rate": 1.8670370370370372e-05, |
| "loss": 1.1609, |
| "mean_token_accuracy": 0.7130995113402605, |
| "num_tokens": 18765324.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 1.0855824522674085, |
| "epoch": 1.552, |
| "grad_norm": 2.4375, |
| "learning_rate": 1.8633333333333333e-05, |
| "loss": 1.1364, |
| "mean_token_accuracy": 0.7168306458741427, |
| "num_tokens": 18966382.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 1.076863781735301, |
| "epoch": 1.568, |
| "grad_norm": 2.5, |
| "learning_rate": 1.85962962962963e-05, |
| "loss": 1.142, |
| "mean_token_accuracy": 0.7190350815653801, |
| "num_tokens": 19160509.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 1.0745064564049245, |
| "epoch": 1.584, |
| "grad_norm": 2.53125, |
| "learning_rate": 1.855925925925926e-05, |
| "loss": 1.1417, |
| "mean_token_accuracy": 0.7172469474375248, |
| "num_tokens": 19354662.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 1.1122350420802831, |
| "epoch": 1.6, |
| "grad_norm": 2.34375, |
| "learning_rate": 1.8522222222222225e-05, |
| "loss": 1.1734, |
| "mean_token_accuracy": 0.7104691743850708, |
| "num_tokens": 19543088.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_biology_entropy": 1.0805462279319764, |
| "eval_biology_loss": 1.3450608253479004, |
| "eval_biology_mean_token_accuracy": 0.6924934167861938, |
| "eval_biology_num_tokens": 19543088.0, |
| "eval_biology_runtime": 27.523, |
| "eval_biology_samples_per_second": 18.167, |
| "eval_biology_steps_per_second": 4.542, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_chemistry_entropy": 0.9468662395477295, |
| "eval_chemistry_loss": 1.3248136043548584, |
| "eval_chemistry_mean_token_accuracy": 0.6985441679954529, |
| "eval_chemistry_num_tokens": 19543088.0, |
| "eval_chemistry_runtime": 34.5201, |
| "eval_chemistry_samples_per_second": 14.484, |
| "eval_chemistry_steps_per_second": 3.621, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_math_entropy": 1.1085859837532044, |
| "eval_math_loss": 1.9135043621063232, |
| "eval_math_mean_token_accuracy": 0.6110133183002472, |
| "eval_math_num_tokens": 19543088.0, |
| "eval_math_runtime": 35.6871, |
| "eval_math_samples_per_second": 14.011, |
| "eval_math_steps_per_second": 3.503, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_physics_entropy": 1.0443167657852173, |
| "eval_physics_loss": 1.470659613609314, |
| "eval_physics_mean_token_accuracy": 0.6771164288520813, |
| "eval_physics_num_tokens": 19543088.0, |
| "eval_physics_runtime": 41.5025, |
| "eval_physics_samples_per_second": 12.047, |
| "eval_physics_steps_per_second": 3.012, |
| "step": 1000 |
| }, |
| { |
| "entropy": 1.0846907667815686, |
| "epoch": 1.616, |
| "grad_norm": 2.375, |
| "learning_rate": 1.8485185185185186e-05, |
| "loss": 1.1398, |
| "mean_token_accuracy": 0.7176602650433779, |
| "num_tokens": 19743506.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 1.1138011269271373, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 2.6875, |
| "learning_rate": 1.844814814814815e-05, |
| "loss": 1.1748, |
| "mean_token_accuracy": 0.7116489164531231, |
| "num_tokens": 19937567.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 1.1001557804644109, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 2.546875, |
| "learning_rate": 1.8411111111111113e-05, |
| "loss": 1.161, |
| "mean_token_accuracy": 0.7135421529412269, |
| "num_tokens": 20129743.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 1.0677352584898472, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 2.34375, |
| "learning_rate": 1.8374074074074074e-05, |
| "loss": 1.1328, |
| "mean_token_accuracy": 0.719743974506855, |
| "num_tokens": 20331548.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 1.099223655089736, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 3.1875, |
| "learning_rate": 1.8337037037037036e-05, |
| "loss": 1.1616, |
| "mean_token_accuracy": 0.7141391549259424, |
| "num_tokens": 20524247.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 1.09108783043921, |
| "epoch": 1.696, |
| "grad_norm": 2.265625, |
| "learning_rate": 1.83e-05, |
| "loss": 1.1479, |
| "mean_token_accuracy": 0.7167260859161615, |
| "num_tokens": 20718121.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 1.1031555369496346, |
| "epoch": 1.712, |
| "grad_norm": 2.515625, |
| "learning_rate": 1.8262962962962966e-05, |
| "loss": 1.1619, |
| "mean_token_accuracy": 0.7138700798153877, |
| "num_tokens": 20921419.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 1.1019545067101717, |
| "epoch": 1.728, |
| "grad_norm": 2.453125, |
| "learning_rate": 1.8225925925925927e-05, |
| "loss": 1.1527, |
| "mean_token_accuracy": 0.7123929090797901, |
| "num_tokens": 21118581.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 1.0738500457257032, |
| "epoch": 1.744, |
| "grad_norm": 2.5625, |
| "learning_rate": 1.8188888888888892e-05, |
| "loss": 1.1274, |
| "mean_token_accuracy": 0.7197212137281894, |
| "num_tokens": 21315128.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 1.0915827210992575, |
| "epoch": 1.76, |
| "grad_norm": 2.703125, |
| "learning_rate": 1.8151851851851854e-05, |
| "loss": 1.1566, |
| "mean_token_accuracy": 0.715040273591876, |
| "num_tokens": 21511066.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_biology_entropy": 1.0931137771606445, |
| "eval_biology_loss": 1.3361490964889526, |
| "eval_biology_mean_token_accuracy": 0.6935926580429077, |
| "eval_biology_num_tokens": 21511066.0, |
| "eval_biology_runtime": 27.5354, |
| "eval_biology_samples_per_second": 18.158, |
| "eval_biology_steps_per_second": 4.54, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_chemistry_entropy": 0.949874951839447, |
| "eval_chemistry_loss": 1.3176461458206177, |
| "eval_chemistry_mean_token_accuracy": 0.6990127840042114, |
| "eval_chemistry_num_tokens": 21511066.0, |
| "eval_chemistry_runtime": 34.8248, |
| "eval_chemistry_samples_per_second": 14.358, |
| "eval_chemistry_steps_per_second": 3.589, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_math_entropy": 1.0966920719146729, |
| "eval_math_loss": 1.8695012331008911, |
| "eval_math_mean_token_accuracy": 0.6181940336227417, |
| "eval_math_num_tokens": 21511066.0, |
| "eval_math_runtime": 35.8392, |
| "eval_math_samples_per_second": 13.951, |
| "eval_math_steps_per_second": 3.488, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_physics_entropy": 1.039449851512909, |
| "eval_physics_loss": 1.4676064252853394, |
| "eval_physics_mean_token_accuracy": 0.6776353507041931, |
| "eval_physics_num_tokens": 21511066.0, |
| "eval_physics_runtime": 41.5476, |
| "eval_physics_samples_per_second": 12.034, |
| "eval_physics_steps_per_second": 3.009, |
| "step": 1100 |
| }, |
| { |
| "entropy": 1.1009868506342173, |
| "epoch": 1.776, |
| "grad_norm": 2.390625, |
| "learning_rate": 1.8114814814814816e-05, |
| "loss": 1.1625, |
| "mean_token_accuracy": 0.7126704059541226, |
| "num_tokens": 21704368.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 1.0688580960035323, |
| "epoch": 1.792, |
| "grad_norm": 2.515625, |
| "learning_rate": 1.8077777777777777e-05, |
| "loss": 1.1255, |
| "mean_token_accuracy": 0.7196867290884257, |
| "num_tokens": 21905784.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 1.0956027507781982, |
| "epoch": 1.808, |
| "grad_norm": 2.78125, |
| "learning_rate": 1.8040740740740742e-05, |
| "loss": 1.1623, |
| "mean_token_accuracy": 0.7157878085970879, |
| "num_tokens": 22094159.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 1.1225024968385697, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 2.5625, |
| "learning_rate": 1.8003703703703707e-05, |
| "loss": 1.169, |
| "mean_token_accuracy": 0.7094971932470798, |
| "num_tokens": 22290654.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 1.1148381125181914, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 2.546875, |
| "learning_rate": 1.796666666666667e-05, |
| "loss": 1.1832, |
| "mean_token_accuracy": 0.708811754733324, |
| "num_tokens": 22485675.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 1.0807621087878942, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 2.671875, |
| "learning_rate": 1.792962962962963e-05, |
| "loss": 1.1454, |
| "mean_token_accuracy": 0.7151207286864519, |
| "num_tokens": 22689900.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 1.0747102357447147, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 2.28125, |
| "learning_rate": 1.7892592592592595e-05, |
| "loss": 1.1325, |
| "mean_token_accuracy": 0.7207924500107765, |
| "num_tokens": 22887275.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 1.1254734165966511, |
| "epoch": 1.888, |
| "grad_norm": 2.875, |
| "learning_rate": 1.7855555555555557e-05, |
| "loss": 1.1869, |
| "mean_token_accuracy": 0.7088476780802011, |
| "num_tokens": 23081241.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 1.0956426173448564, |
| "epoch": 1.904, |
| "grad_norm": 2.5625, |
| "learning_rate": 1.7818518518518518e-05, |
| "loss": 1.1668, |
| "mean_token_accuracy": 0.7139182474464179, |
| "num_tokens": 23276417.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 1.1328827075660228, |
| "epoch": 1.92, |
| "grad_norm": 2.609375, |
| "learning_rate": 1.7781481481481483e-05, |
| "loss": 1.177, |
| "mean_token_accuracy": 0.7098524257540703, |
| "num_tokens": 23476340.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_biology_entropy": 1.0637187623977662, |
| "eval_biology_loss": 1.330007791519165, |
| "eval_biology_mean_token_accuracy": 0.6948464093208313, |
| "eval_biology_num_tokens": 23476340.0, |
| "eval_biology_runtime": 27.5667, |
| "eval_biology_samples_per_second": 18.138, |
| "eval_biology_steps_per_second": 4.534, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_chemistry_entropy": 0.9230625042915345, |
| "eval_chemistry_loss": 1.303159475326538, |
| "eval_chemistry_mean_token_accuracy": 0.7019083909988403, |
| "eval_chemistry_num_tokens": 23476340.0, |
| "eval_chemistry_runtime": 34.8108, |
| "eval_chemistry_samples_per_second": 14.363, |
| "eval_chemistry_steps_per_second": 3.591, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_math_entropy": 1.0601022329330445, |
| "eval_math_loss": 1.8581321239471436, |
| "eval_math_mean_token_accuracy": 0.6215305128097535, |
| "eval_math_num_tokens": 23476340.0, |
| "eval_math_runtime": 35.8178, |
| "eval_math_samples_per_second": 13.96, |
| "eval_math_steps_per_second": 3.49, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_physics_entropy": 1.0067597246170044, |
| "eval_physics_loss": 1.4505807161331177, |
| "eval_physics_mean_token_accuracy": 0.6818430233001709, |
| "eval_physics_num_tokens": 23476340.0, |
| "eval_physics_runtime": 41.5079, |
| "eval_physics_samples_per_second": 12.046, |
| "eval_physics_steps_per_second": 3.011, |
| "step": 1200 |
| }, |
| { |
| "entropy": 1.0840563517063857, |
| "epoch": 1.936, |
| "grad_norm": 2.453125, |
| "learning_rate": 1.7744444444444445e-05, |
| "loss": 1.1373, |
| "mean_token_accuracy": 0.7173909548670053, |
| "num_tokens": 23673208.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 1.1050400588661433, |
| "epoch": 1.952, |
| "grad_norm": 2.484375, |
| "learning_rate": 1.770740740740741e-05, |
| "loss": 1.1572, |
| "mean_token_accuracy": 0.7145154111087322, |
| "num_tokens": 23870993.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 1.109605184942484, |
| "epoch": 1.968, |
| "grad_norm": 2.53125, |
| "learning_rate": 1.767037037037037e-05, |
| "loss": 1.1846, |
| "mean_token_accuracy": 0.7109895326197148, |
| "num_tokens": 24066234.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 1.1098197661340237, |
| "epoch": 1.984, |
| "grad_norm": 2.328125, |
| "learning_rate": 1.7633333333333336e-05, |
| "loss": 1.1677, |
| "mean_token_accuracy": 0.7122665207833052, |
| "num_tokens": 24262389.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 1.1092739596962928, |
| "epoch": 2.0, |
| "grad_norm": 2.5625, |
| "learning_rate": 1.7596296296296298e-05, |
| "loss": 1.1781, |
| "mean_token_accuracy": 0.713566455245018, |
| "num_tokens": 24455544.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 1.0043137110769749, |
| "epoch": 2.016, |
| "grad_norm": 2.765625, |
| "learning_rate": 1.755925925925926e-05, |
| "loss": 1.0483, |
| "mean_token_accuracy": 0.7371486712247133, |
| "num_tokens": 24645425.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 1.0089781265705824, |
| "epoch": 2.032, |
| "grad_norm": 2.703125, |
| "learning_rate": 1.7522222222222224e-05, |
| "loss": 1.0577, |
| "mean_token_accuracy": 0.7354639135301113, |
| "num_tokens": 24835956.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.9756499044597149, |
| "epoch": 2.048, |
| "grad_norm": 2.765625, |
| "learning_rate": 1.7485185185185186e-05, |
| "loss": 1.0234, |
| "mean_token_accuracy": 0.7406638573855162, |
| "num_tokens": 25036470.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.9923272367566824, |
| "epoch": 2.064, |
| "grad_norm": 2.59375, |
| "learning_rate": 1.744814814814815e-05, |
| "loss": 1.0429, |
| "mean_token_accuracy": 0.7380331251770258, |
| "num_tokens": 25230959.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.9767786219716073, |
| "epoch": 2.08, |
| "grad_norm": 2.8125, |
| "learning_rate": 1.7411111111111112e-05, |
| "loss": 1.0397, |
| "mean_token_accuracy": 0.7391406297683716, |
| "num_tokens": 25422625.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_biology_entropy": 0.9924388265609742, |
| "eval_biology_loss": 1.3691943883895874, |
| "eval_biology_mean_token_accuracy": 0.6921270470619202, |
| "eval_biology_num_tokens": 25422625.0, |
| "eval_biology_runtime": 27.5226, |
| "eval_biology_samples_per_second": 18.167, |
| "eval_biology_steps_per_second": 4.542, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_chemistry_entropy": 0.8449653520584106, |
| "eval_chemistry_loss": 1.3724958896636963, |
| "eval_chemistry_mean_token_accuracy": 0.6972775468826294, |
| "eval_chemistry_num_tokens": 25422625.0, |
| "eval_chemistry_runtime": 34.7815, |
| "eval_chemistry_samples_per_second": 14.375, |
| "eval_chemistry_steps_per_second": 3.594, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_math_entropy": 0.9701335663795472, |
| "eval_math_loss": 1.979202389717102, |
| "eval_math_mean_token_accuracy": 0.616259946346283, |
| "eval_math_num_tokens": 25422625.0, |
| "eval_math_runtime": 35.839, |
| "eval_math_samples_per_second": 13.951, |
| "eval_math_steps_per_second": 3.488, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_physics_entropy": 0.9231480684280395, |
| "eval_physics_loss": 1.533275842666626, |
| "eval_physics_mean_token_accuracy": 0.6762914519309997, |
| "eval_physics_num_tokens": 25422625.0, |
| "eval_physics_runtime": 41.5395, |
| "eval_physics_samples_per_second": 12.037, |
| "eval_physics_steps_per_second": 3.009, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.9958227720111609, |
| "epoch": 2.096, |
| "grad_norm": 2.640625, |
| "learning_rate": 1.7374074074074077e-05, |
| "loss": 1.0489, |
| "mean_token_accuracy": 0.7380556058138609, |
| "num_tokens": 25614286.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.9596387047320605, |
| "epoch": 2.112, |
| "grad_norm": 2.53125, |
| "learning_rate": 1.733703703703704e-05, |
| "loss": 1.0117, |
| "mean_token_accuracy": 0.7437543950974941, |
| "num_tokens": 25810384.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.9700033210217953, |
| "epoch": 2.128, |
| "grad_norm": 2.703125, |
| "learning_rate": 1.73e-05, |
| "loss": 1.0293, |
| "mean_token_accuracy": 0.7417288426309824, |
| "num_tokens": 26005047.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.9836055677384138, |
| "epoch": 2.144, |
| "grad_norm": 2.984375, |
| "learning_rate": 1.7262962962962962e-05, |
| "loss": 1.0385, |
| "mean_token_accuracy": 0.7380954034626483, |
| "num_tokens": 26207169.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.9896935578435659, |
| "epoch": 2.16, |
| "grad_norm": 2.6875, |
| "learning_rate": 1.7225925925925927e-05, |
| "loss": 1.0443, |
| "mean_token_accuracy": 0.7387308221310377, |
| "num_tokens": 26402150.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 1.0001810498535633, |
| "epoch": 2.176, |
| "grad_norm": 2.671875, |
| "learning_rate": 1.7188888888888892e-05, |
| "loss": 1.0565, |
| "mean_token_accuracy": 0.7334125578403473, |
| "num_tokens": 26599464.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.9603279117494822, |
| "epoch": 2.192, |
| "grad_norm": 2.40625, |
| "learning_rate": 1.7151851851851853e-05, |
| "loss": 1.016, |
| "mean_token_accuracy": 0.7426716230809689, |
| "num_tokens": 26790601.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.9929966073483228, |
| "epoch": 2.208, |
| "grad_norm": 2.71875, |
| "learning_rate": 1.7114814814814815e-05, |
| "loss": 1.0446, |
| "mean_token_accuracy": 0.7385065902024508, |
| "num_tokens": 26988598.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 1.0099530704319477, |
| "epoch": 2.224, |
| "grad_norm": 2.59375, |
| "learning_rate": 1.707777777777778e-05, |
| "loss": 1.0682, |
| "mean_token_accuracy": 0.7335593592375517, |
| "num_tokens": 27181933.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.96002363152802, |
| "epoch": 2.24, |
| "grad_norm": 2.578125, |
| "learning_rate": 1.704074074074074e-05, |
| "loss": 1.016, |
| "mean_token_accuracy": 0.7427360139787197, |
| "num_tokens": 27378686.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_biology_entropy": 0.9901858034133911, |
| "eval_biology_loss": 1.3687505722045898, |
| "eval_biology_mean_token_accuracy": 0.6931924157142639, |
| "eval_biology_num_tokens": 27378686.0, |
| "eval_biology_runtime": 27.5158, |
| "eval_biology_samples_per_second": 18.171, |
| "eval_biology_steps_per_second": 4.543, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_chemistry_entropy": 0.8399234094619751, |
| "eval_chemistry_loss": 1.377020239830017, |
| "eval_chemistry_mean_token_accuracy": 0.6967816543579102, |
| "eval_chemistry_num_tokens": 27378686.0, |
| "eval_chemistry_runtime": 34.8057, |
| "eval_chemistry_samples_per_second": 14.365, |
| "eval_chemistry_steps_per_second": 3.591, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_math_entropy": 0.9545157217979431, |
| "eval_math_loss": 2.0055806636810303, |
| "eval_math_mean_token_accuracy": 0.6127783393859864, |
| "eval_math_num_tokens": 27378686.0, |
| "eval_math_runtime": 35.8347, |
| "eval_math_samples_per_second": 13.953, |
| "eval_math_steps_per_second": 3.488, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_physics_entropy": 0.9126699995994568, |
| "eval_physics_loss": 1.5412017107009888, |
| "eval_physics_mean_token_accuracy": 0.6753737378120422, |
| "eval_physics_num_tokens": 27378686.0, |
| "eval_physics_runtime": 41.5129, |
| "eval_physics_samples_per_second": 12.044, |
| "eval_physics_steps_per_second": 3.011, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.9692367669194937, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 2.671875, |
| "learning_rate": 1.7003703703703703e-05, |
| "loss": 1.032, |
| "mean_token_accuracy": 0.7413300447165966, |
| "num_tokens": 27571908.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.962577823549509, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 2.625, |
| "learning_rate": 1.6966666666666668e-05, |
| "loss": 1.019, |
| "mean_token_accuracy": 0.7413133516907692, |
| "num_tokens": 27770478.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 1.006012573838234, |
| "epoch": 2.288, |
| "grad_norm": 2.671875, |
| "learning_rate": 1.6929629629629633e-05, |
| "loss": 1.06, |
| "mean_token_accuracy": 0.7331223022192717, |
| "num_tokens": 27965558.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.9969842620193958, |
| "epoch": 2.304, |
| "grad_norm": 2.578125, |
| "learning_rate": 1.6892592592592594e-05, |
| "loss": 1.0568, |
| "mean_token_accuracy": 0.734639760479331, |
| "num_tokens": 28165624.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.972690061852336, |
| "epoch": 2.32, |
| "grad_norm": 2.65625, |
| "learning_rate": 1.6855555555555556e-05, |
| "loss": 1.0283, |
| "mean_token_accuracy": 0.7417871564626694, |
| "num_tokens": 28363594.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.9879562474787236, |
| "epoch": 2.336, |
| "grad_norm": 2.671875, |
| "learning_rate": 1.681851851851852e-05, |
| "loss": 1.0417, |
| "mean_token_accuracy": 0.7382532250136137, |
| "num_tokens": 28558104.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 1.0057890437543393, |
| "epoch": 2.352, |
| "grad_norm": 2.5625, |
| "learning_rate": 1.6781481481481483e-05, |
| "loss": 1.0637, |
| "mean_token_accuracy": 0.7329543225467205, |
| "num_tokens": 28747265.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.9955759733915329, |
| "epoch": 2.368, |
| "grad_norm": 2.84375, |
| "learning_rate": 1.6744444444444444e-05, |
| "loss": 1.0421, |
| "mean_token_accuracy": 0.7366544388234615, |
| "num_tokens": 28944694.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 1.0063096433877945, |
| "epoch": 2.384, |
| "grad_norm": 2.75, |
| "learning_rate": 1.670740740740741e-05, |
| "loss": 1.0606, |
| "mean_token_accuracy": 0.7337229486554862, |
| "num_tokens": 29136617.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.9919678378850222, |
| "epoch": 2.4, |
| "grad_norm": 2.6875, |
| "learning_rate": 1.667037037037037e-05, |
| "loss": 1.0426, |
| "mean_token_accuracy": 0.7359879236668349, |
| "num_tokens": 29333526.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_biology_entropy": 0.9798638577461243, |
| "eval_biology_loss": 1.3690063953399658, |
| "eval_biology_mean_token_accuracy": 0.6930591049194336, |
| "eval_biology_num_tokens": 29333526.0, |
| "eval_biology_runtime": 27.5545, |
| "eval_biology_samples_per_second": 18.146, |
| "eval_biology_steps_per_second": 4.536, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_chemistry_entropy": 0.8374483752250671, |
| "eval_chemistry_loss": 1.3725800514221191, |
| "eval_chemistry_mean_token_accuracy": 0.6977551965713501, |
| "eval_chemistry_num_tokens": 29333526.0, |
| "eval_chemistry_runtime": 34.5244, |
| "eval_chemistry_samples_per_second": 14.483, |
| "eval_chemistry_steps_per_second": 3.621, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_math_entropy": 0.9796739134788514, |
| "eval_math_loss": 1.9708421230316162, |
| "eval_math_mean_token_accuracy": 0.6159570138454438, |
| "eval_math_num_tokens": 29333526.0, |
| "eval_math_runtime": 35.7215, |
| "eval_math_samples_per_second": 13.997, |
| "eval_math_steps_per_second": 3.499, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_physics_entropy": 0.9184642505645751, |
| "eval_physics_loss": 1.532168984413147, |
| "eval_physics_mean_token_accuracy": 0.6767212514877319, |
| "eval_physics_num_tokens": 29333526.0, |
| "eval_physics_runtime": 41.4698, |
| "eval_physics_samples_per_second": 12.057, |
| "eval_physics_steps_per_second": 3.014, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.9870230372995138, |
| "epoch": 2.416, |
| "grad_norm": 2.71875, |
| "learning_rate": 1.6633333333333336e-05, |
| "loss": 1.0479, |
| "mean_token_accuracy": 0.7350806698203087, |
| "num_tokens": 29534153.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.9857139840722084, |
| "epoch": 2.432, |
| "grad_norm": 2.71875, |
| "learning_rate": 1.6596296296296297e-05, |
| "loss": 1.0372, |
| "mean_token_accuracy": 0.7372451640665532, |
| "num_tokens": 29736571.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 1.0195895943790674, |
| "epoch": 2.448, |
| "grad_norm": 3.03125, |
| "learning_rate": 1.6559259259259262e-05, |
| "loss": 1.0802, |
| "mean_token_accuracy": 0.7314553778618574, |
| "num_tokens": 29932423.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.9653120532631874, |
| "epoch": 2.464, |
| "grad_norm": 2.65625, |
| "learning_rate": 1.6522222222222224e-05, |
| "loss": 1.0129, |
| "mean_token_accuracy": 0.7418248776346446, |
| "num_tokens": 30137125.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.9762739170342684, |
| "epoch": 2.48, |
| "grad_norm": 2.578125, |
| "learning_rate": 1.6485185185185185e-05, |
| "loss": 1.0315, |
| "mean_token_accuracy": 0.738971047103405, |
| "num_tokens": 30341656.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.995275142416358, |
| "epoch": 2.496, |
| "grad_norm": 2.546875, |
| "learning_rate": 1.644814814814815e-05, |
| "loss": 1.0448, |
| "mean_token_accuracy": 0.73451181165874, |
| "num_tokens": 30533732.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 1.0058202154934406, |
| "epoch": 2.512, |
| "grad_norm": 3.140625, |
| "learning_rate": 1.6411111111111112e-05, |
| "loss": 1.0681, |
| "mean_token_accuracy": 0.7326466295868158, |
| "num_tokens": 30730599.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 1.0007912684231997, |
| "epoch": 2.528, |
| "grad_norm": 2.734375, |
| "learning_rate": 1.6374074074074077e-05, |
| "loss": 1.052, |
| "mean_token_accuracy": 0.7349742811173201, |
| "num_tokens": 30927950.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.9980495974421502, |
| "epoch": 2.544, |
| "grad_norm": 2.59375, |
| "learning_rate": 1.6337037037037038e-05, |
| "loss": 1.0633, |
| "mean_token_accuracy": 0.7333921913057566, |
| "num_tokens": 31125667.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 1.0160194922238588, |
| "epoch": 2.56, |
| "grad_norm": 2.859375, |
| "learning_rate": 1.63e-05, |
| "loss": 1.0703, |
| "mean_token_accuracy": 0.7330452054738998, |
| "num_tokens": 31315742.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_biology_entropy": 0.9697687101364135, |
| "eval_biology_loss": 1.363986611366272, |
| "eval_biology_mean_token_accuracy": 0.6941581168174744, |
| "eval_biology_num_tokens": 31315742.0, |
| "eval_biology_runtime": 27.5772, |
| "eval_biology_samples_per_second": 18.131, |
| "eval_biology_steps_per_second": 4.533, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_chemistry_entropy": 0.8340853586196899, |
| "eval_chemistry_loss": 1.3694249391555786, |
| "eval_chemistry_mean_token_accuracy": 0.6980492143630982, |
| "eval_chemistry_num_tokens": 31315742.0, |
| "eval_chemistry_runtime": 34.8138, |
| "eval_chemistry_samples_per_second": 14.362, |
| "eval_chemistry_steps_per_second": 3.591, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_math_entropy": 0.9793781757354736, |
| "eval_math_loss": 1.9548852443695068, |
| "eval_math_mean_token_accuracy": 0.6173824925422668, |
| "eval_math_num_tokens": 31315742.0, |
| "eval_math_runtime": 35.8514, |
| "eval_math_samples_per_second": 13.946, |
| "eval_math_steps_per_second": 3.487, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_physics_entropy": 0.9162781934738159, |
| "eval_physics_loss": 1.5209226608276367, |
| "eval_physics_mean_token_accuracy": 0.6783299732208252, |
| "eval_physics_num_tokens": 31315742.0, |
| "eval_physics_runtime": 41.6037, |
| "eval_physics_samples_per_second": 12.018, |
| "eval_physics_steps_per_second": 3.005, |
| "step": 1600 |
| }, |
| { |
| "entropy": 1.0157440803945064, |
| "epoch": 2.576, |
| "grad_norm": 2.53125, |
| "learning_rate": 1.6262962962962965e-05, |
| "loss": 1.0784, |
| "mean_token_accuracy": 0.732684288173914, |
| "num_tokens": 31505374.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 1.0260655630379916, |
| "epoch": 2.592, |
| "grad_norm": 2.703125, |
| "learning_rate": 1.6225925925925926e-05, |
| "loss": 1.0839, |
| "mean_token_accuracy": 0.7257251992821694, |
| "num_tokens": 31701680.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 1.0221612501889468, |
| "epoch": 2.608, |
| "grad_norm": 2.671875, |
| "learning_rate": 1.6188888888888888e-05, |
| "loss": 1.0764, |
| "mean_token_accuracy": 0.7304493173956871, |
| "num_tokens": 31893221.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.993102552369237, |
| "epoch": 2.624, |
| "grad_norm": 2.671875, |
| "learning_rate": 1.6151851851851853e-05, |
| "loss": 1.0478, |
| "mean_token_accuracy": 0.7348294869065285, |
| "num_tokens": 32087176.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 1.0137774087488651, |
| "epoch": 2.64, |
| "grad_norm": 2.578125, |
| "learning_rate": 1.6114814814814818e-05, |
| "loss": 1.0647, |
| "mean_token_accuracy": 0.7330294415354729, |
| "num_tokens": 32280689.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.9745670046657324, |
| "epoch": 2.656, |
| "grad_norm": 2.6875, |
| "learning_rate": 1.607777777777778e-05, |
| "loss": 1.0368, |
| "mean_token_accuracy": 0.7391632381826639, |
| "num_tokens": 32483392.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.9923187121748924, |
| "epoch": 2.672, |
| "grad_norm": 2.5625, |
| "learning_rate": 1.604074074074074e-05, |
| "loss": 1.0482, |
| "mean_token_accuracy": 0.7355044964700937, |
| "num_tokens": 32683246.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.9856970235705376, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 2.625, |
| "learning_rate": 1.6003703703703706e-05, |
| "loss": 1.0355, |
| "mean_token_accuracy": 0.7383067451417447, |
| "num_tokens": 32884993.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.9969516977667808, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 2.59375, |
| "learning_rate": 1.5966666666666667e-05, |
| "loss": 1.0462, |
| "mean_token_accuracy": 0.7363014262169599, |
| "num_tokens": 33088388.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.988432103022933, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 2.890625, |
| "learning_rate": 1.592962962962963e-05, |
| "loss": 1.0476, |
| "mean_token_accuracy": 0.7358232375234366, |
| "num_tokens": 33287208.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_biology_entropy": 0.9908420953750611, |
| "eval_biology_loss": 1.3577314615249634, |
| "eval_biology_mean_token_accuracy": 0.6939261827468872, |
| "eval_biology_num_tokens": 33287208.0, |
| "eval_biology_runtime": 27.5386, |
| "eval_biology_samples_per_second": 18.156, |
| "eval_biology_steps_per_second": 4.539, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_chemistry_entropy": 0.8477892417907715, |
| "eval_chemistry_loss": 1.3660188913345337, |
| "eval_chemistry_mean_token_accuracy": 0.6991322560310363, |
| "eval_chemistry_num_tokens": 33287208.0, |
| "eval_chemistry_runtime": 34.823, |
| "eval_chemistry_samples_per_second": 14.358, |
| "eval_chemistry_steps_per_second": 3.59, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_math_entropy": 0.9799207682609559, |
| "eval_math_loss": 1.9764255285263062, |
| "eval_math_mean_token_accuracy": 0.6149459352493286, |
| "eval_math_num_tokens": 33287208.0, |
| "eval_math_runtime": 35.8703, |
| "eval_math_samples_per_second": 13.939, |
| "eval_math_steps_per_second": 3.485, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_physics_entropy": 0.9246280765533448, |
| "eval_physics_loss": 1.5206706523895264, |
| "eval_physics_mean_token_accuracy": 0.6784826369285584, |
| "eval_physics_num_tokens": 33287208.0, |
| "eval_physics_runtime": 41.6175, |
| "eval_physics_samples_per_second": 12.014, |
| "eval_physics_steps_per_second": 3.004, |
| "step": 1700 |
| }, |
| { |
| "entropy": 1.015086841583252, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 2.59375, |
| "learning_rate": 1.5892592592592594e-05, |
| "loss": 1.0783, |
| "mean_token_accuracy": 0.730317522212863, |
| "num_tokens": 33474750.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 1.0128032747656106, |
| "epoch": 2.752, |
| "grad_norm": 2.75, |
| "learning_rate": 1.585555555555556e-05, |
| "loss": 1.0705, |
| "mean_token_accuracy": 0.731030648574233, |
| "num_tokens": 33664302.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 1.030053498968482, |
| "epoch": 2.768, |
| "grad_norm": 2.578125, |
| "learning_rate": 1.581851851851852e-05, |
| "loss": 1.0851, |
| "mean_token_accuracy": 0.7275784995406867, |
| "num_tokens": 33854081.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 1.0005322974175215, |
| "epoch": 2.784, |
| "grad_norm": 2.484375, |
| "learning_rate": 1.5781481481481482e-05, |
| "loss": 1.0622, |
| "mean_token_accuracy": 0.7345380343496799, |
| "num_tokens": 34053149.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 1.0030489567667247, |
| "epoch": 2.8, |
| "grad_norm": 2.65625, |
| "learning_rate": 1.5744444444444447e-05, |
| "loss": 1.0571, |
| "mean_token_accuracy": 0.7348426088690758, |
| "num_tokens": 34254105.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.992670550569892, |
| "epoch": 2.816, |
| "grad_norm": 2.5, |
| "learning_rate": 1.570740740740741e-05, |
| "loss": 1.0477, |
| "mean_token_accuracy": 0.7356164410710335, |
| "num_tokens": 34453342.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 1.0293358892202378, |
| "epoch": 2.832, |
| "grad_norm": 2.640625, |
| "learning_rate": 1.567037037037037e-05, |
| "loss": 1.0813, |
| "mean_token_accuracy": 0.7298110734671355, |
| "num_tokens": 34644113.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 1.013759650476277, |
| "epoch": 2.848, |
| "grad_norm": 2.671875, |
| "learning_rate": 1.5633333333333335e-05, |
| "loss": 1.0761, |
| "mean_token_accuracy": 0.7321403097361326, |
| "num_tokens": 34835062.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 1.0077157508581878, |
| "epoch": 2.864, |
| "grad_norm": 2.734375, |
| "learning_rate": 1.55962962962963e-05, |
| "loss": 1.0664, |
| "mean_token_accuracy": 0.731912524625659, |
| "num_tokens": 35032361.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 1.0294968254864216, |
| "epoch": 2.88, |
| "grad_norm": 2.765625, |
| "learning_rate": 1.555925925925926e-05, |
| "loss": 1.0826, |
| "mean_token_accuracy": 0.7297796905040741, |
| "num_tokens": 35226653.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_biology_entropy": 0.9937820811271667, |
| "eval_biology_loss": 1.353894829750061, |
| "eval_biology_mean_token_accuracy": 0.6938789577484131, |
| "eval_biology_num_tokens": 35226653.0, |
| "eval_biology_runtime": 27.5559, |
| "eval_biology_samples_per_second": 18.145, |
| "eval_biology_steps_per_second": 4.536, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_chemistry_entropy": 0.8393283033370972, |
| "eval_chemistry_loss": 1.3614035844802856, |
| "eval_chemistry_mean_token_accuracy": 0.6999081645011902, |
| "eval_chemistry_num_tokens": 35226653.0, |
| "eval_chemistry_runtime": 34.8155, |
| "eval_chemistry_samples_per_second": 14.361, |
| "eval_chemistry_steps_per_second": 3.59, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_math_entropy": 0.958376314163208, |
| "eval_math_loss": 1.9908337593078613, |
| "eval_math_mean_token_accuracy": 0.6151066827774048, |
| "eval_math_num_tokens": 35226653.0, |
| "eval_math_runtime": 35.8758, |
| "eval_math_samples_per_second": 13.937, |
| "eval_math_steps_per_second": 3.484, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_physics_entropy": 0.9136257534027099, |
| "eval_physics_loss": 1.519931435585022, |
| "eval_physics_mean_token_accuracy": 0.6788474645614624, |
| "eval_physics_num_tokens": 35226653.0, |
| "eval_physics_runtime": 41.5999, |
| "eval_physics_samples_per_second": 12.019, |
| "eval_physics_steps_per_second": 3.005, |
| "step": 1800 |
| }, |
| { |
| "entropy": 1.0241970747709275, |
| "epoch": 2.896, |
| "grad_norm": 2.625, |
| "learning_rate": 1.5522222222222223e-05, |
| "loss": 1.0738, |
| "mean_token_accuracy": 0.7299285631626844, |
| "num_tokens": 35421873.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 1.0258236292749643, |
| "epoch": 2.912, |
| "grad_norm": 2.734375, |
| "learning_rate": 1.5485185185185188e-05, |
| "loss": 1.0883, |
| "mean_token_accuracy": 0.7278658114373684, |
| "num_tokens": 35614590.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 1.0190716098994017, |
| "epoch": 2.928, |
| "grad_norm": 2.8125, |
| "learning_rate": 1.544814814814815e-05, |
| "loss": 1.0843, |
| "mean_token_accuracy": 0.7294420916587114, |
| "num_tokens": 35806494.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 1.019536604732275, |
| "epoch": 2.944, |
| "grad_norm": 2.75, |
| "learning_rate": 1.541111111111111e-05, |
| "loss": 1.0773, |
| "mean_token_accuracy": 0.7327182784676551, |
| "num_tokens": 36001867.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.9926906757056713, |
| "epoch": 2.96, |
| "grad_norm": 2.796875, |
| "learning_rate": 1.5374074074074076e-05, |
| "loss": 1.0497, |
| "mean_token_accuracy": 0.7346879154443741, |
| "num_tokens": 36198806.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 1.0060346502810718, |
| "epoch": 2.976, |
| "grad_norm": 2.625, |
| "learning_rate": 1.5337037037037038e-05, |
| "loss": 1.0625, |
| "mean_token_accuracy": 0.7337981924414635, |
| "num_tokens": 36393098.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 1.018786373361945, |
| "epoch": 2.992, |
| "grad_norm": 2.796875, |
| "learning_rate": 1.5300000000000003e-05, |
| "loss": 1.0767, |
| "mean_token_accuracy": 0.7307282339781522, |
| "num_tokens": 36589074.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 1.0170041617006063, |
| "epoch": 3.008, |
| "grad_norm": 2.734375, |
| "learning_rate": 1.5262962962962964e-05, |
| "loss": 1.0491, |
| "mean_token_accuracy": 0.7375065568834543, |
| "num_tokens": 36778710.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.9372840091586113, |
| "epoch": 3.024, |
| "grad_norm": 2.734375, |
| "learning_rate": 1.5225925925925926e-05, |
| "loss": 0.9999, |
| "mean_token_accuracy": 0.746622646600008, |
| "num_tokens": 36966789.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.9152062568813563, |
| "epoch": 3.04, |
| "grad_norm": 2.890625, |
| "learning_rate": 1.518888888888889e-05, |
| "loss": 0.968, |
| "mean_token_accuracy": 0.7552635859698057, |
| "num_tokens": 37156529.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_biology_entropy": 0.9203187170028686, |
| "eval_biology_loss": 1.3961647748947144, |
| "eval_biology_mean_token_accuracy": 0.6918789324760437, |
| "eval_biology_num_tokens": 37156529.0, |
| "eval_biology_runtime": 27.5466, |
| "eval_biology_samples_per_second": 18.151, |
| "eval_biology_steps_per_second": 4.538, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_chemistry_entropy": 0.7799582343101501, |
| "eval_chemistry_loss": 1.4286096096038818, |
| "eval_chemistry_mean_token_accuracy": 0.6953253383636475, |
| "eval_chemistry_num_tokens": 37156529.0, |
| "eval_chemistry_runtime": 34.8301, |
| "eval_chemistry_samples_per_second": 14.355, |
| "eval_chemistry_steps_per_second": 3.589, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_math_entropy": 0.8959285621643066, |
| "eval_math_loss": 2.0766634941101074, |
| "eval_math_mean_token_accuracy": 0.6109981193542481, |
| "eval_math_num_tokens": 37156529.0, |
| "eval_math_runtime": 35.8739, |
| "eval_math_samples_per_second": 13.938, |
| "eval_math_steps_per_second": 3.484, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_physics_entropy": 0.8492022409439087, |
| "eval_physics_loss": 1.5976672172546387, |
| "eval_physics_mean_token_accuracy": 0.6750246415138245, |
| "eval_physics_num_tokens": 37156529.0, |
| "eval_physics_runtime": 41.5867, |
| "eval_physics_samples_per_second": 12.023, |
| "eval_physics_steps_per_second": 3.006, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.9346861682832242, |
| "epoch": 3.056, |
| "grad_norm": 3.09375, |
| "learning_rate": 1.5151851851851854e-05, |
| "loss": 0.9766, |
| "mean_token_accuracy": 0.7512495912611484, |
| "num_tokens": 37351325.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.898235434293747, |
| "epoch": 3.072, |
| "grad_norm": 2.890625, |
| "learning_rate": 1.5114814814814815e-05, |
| "loss": 0.963, |
| "mean_token_accuracy": 0.7571980815380812, |
| "num_tokens": 37541069.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.9269086364656687, |
| "epoch": 3.088, |
| "grad_norm": 2.8125, |
| "learning_rate": 1.507777777777778e-05, |
| "loss": 0.9747, |
| "mean_token_accuracy": 0.754531666636467, |
| "num_tokens": 37733440.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.9060456406325101, |
| "epoch": 3.104, |
| "grad_norm": 2.75, |
| "learning_rate": 1.5040740740740742e-05, |
| "loss": 0.9558, |
| "mean_token_accuracy": 0.756022872403264, |
| "num_tokens": 37933661.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.9189450412988662, |
| "epoch": 3.12, |
| "grad_norm": 2.75, |
| "learning_rate": 1.5003703703703705e-05, |
| "loss": 0.9754, |
| "mean_token_accuracy": 0.7542413301765919, |
| "num_tokens": 38126905.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.9249624401330948, |
| "epoch": 3.136, |
| "grad_norm": 3.1875, |
| "learning_rate": 1.4966666666666667e-05, |
| "loss": 0.9751, |
| "mean_token_accuracy": 0.7536766722798347, |
| "num_tokens": 38316042.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.9087997317314148, |
| "epoch": 3.152, |
| "grad_norm": 3.0, |
| "learning_rate": 1.4929629629629632e-05, |
| "loss": 0.9584, |
| "mean_token_accuracy": 0.7552351832389832, |
| "num_tokens": 38514212.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.9421753611415624, |
| "epoch": 3.168, |
| "grad_norm": 2.75, |
| "learning_rate": 1.4892592592592593e-05, |
| "loss": 0.9957, |
| "mean_token_accuracy": 0.7481138132512569, |
| "num_tokens": 38708822.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.9290799971669912, |
| "epoch": 3.184, |
| "grad_norm": 2.734375, |
| "learning_rate": 1.4855555555555557e-05, |
| "loss": 0.9855, |
| "mean_token_accuracy": 0.7512500148266554, |
| "num_tokens": 38900708.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.9192495409399271, |
| "epoch": 3.2, |
| "grad_norm": 2.84375, |
| "learning_rate": 1.4818518518518518e-05, |
| "loss": 0.9679, |
| "mean_token_accuracy": 0.7533360481262207, |
| "num_tokens": 39102325.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_biology_entropy": 0.9260811395645142, |
| "eval_biology_loss": 1.3970844745635986, |
| "eval_biology_mean_token_accuracy": 0.6919394407272339, |
| "eval_biology_num_tokens": 39102325.0, |
| "eval_biology_runtime": 27.5579, |
| "eval_biology_samples_per_second": 18.144, |
| "eval_biology_steps_per_second": 4.536, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_chemistry_entropy": 0.7810218653678894, |
| "eval_chemistry_loss": 1.4344403743743896, |
| "eval_chemistry_mean_token_accuracy": 0.6947128176689148, |
| "eval_chemistry_num_tokens": 39102325.0, |
| "eval_chemistry_runtime": 34.5093, |
| "eval_chemistry_samples_per_second": 14.489, |
| "eval_chemistry_steps_per_second": 3.622, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_math_entropy": 0.9098891434669495, |
| "eval_math_loss": 2.063753843307495, |
| "eval_math_mean_token_accuracy": 0.6125479984283447, |
| "eval_math_num_tokens": 39102325.0, |
| "eval_math_runtime": 35.6812, |
| "eval_math_samples_per_second": 14.013, |
| "eval_math_steps_per_second": 3.503, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_physics_entropy": 0.854760775089264, |
| "eval_physics_loss": 1.6023722887039185, |
| "eval_physics_mean_token_accuracy": 0.6737847833633422, |
| "eval_physics_num_tokens": 39102325.0, |
| "eval_physics_runtime": 41.4693, |
| "eval_physics_samples_per_second": 12.057, |
| "eval_physics_steps_per_second": 3.014, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 6000, |
| "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": 3.279038413131436e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
