| { |
| "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": 0.9826594747602939, |
| "epoch": 0.016, |
| "grad_norm": 156.0, |
| "learning_rate": 6.000000000000001e-07, |
| "loss": 33.4095, |
| "mean_token_accuracy": 0.6464727316051722, |
| "num_tokens": 280941.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.0350239120423794, |
| "epoch": 0.032, |
| "grad_norm": 138.0, |
| "learning_rate": 1.2666666666666669e-06, |
| "loss": 33.8115, |
| "mean_token_accuracy": 0.6422434952110052, |
| "num_tokens": 558056.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.0517900373786688, |
| "epoch": 0.048, |
| "grad_norm": 120.0, |
| "learning_rate": 1.9333333333333336e-06, |
| "loss": 32.9081, |
| "mean_token_accuracy": 0.6461116325110197, |
| "num_tokens": 836753.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.1479264814406633, |
| "epoch": 0.064, |
| "grad_norm": 96.5, |
| "learning_rate": 2.6e-06, |
| "loss": 31.9094, |
| "mean_token_accuracy": 0.6435929080471396, |
| "num_tokens": 1126446.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.285454823821783, |
| "epoch": 0.08, |
| "grad_norm": 75.5, |
| "learning_rate": 3.266666666666667e-06, |
| "loss": 29.7071, |
| "mean_token_accuracy": 0.6587587364017964, |
| "num_tokens": 1413596.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.388279514759779, |
| "epoch": 0.096, |
| "grad_norm": 49.0, |
| "learning_rate": 3.9333333333333335e-06, |
| "loss": 26.9094, |
| "mean_token_accuracy": 0.6805908571928739, |
| "num_tokens": 1701193.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.4036601662635804, |
| "epoch": 0.112, |
| "grad_norm": 40.5, |
| "learning_rate": 4.600000000000001e-06, |
| "loss": 24.8593, |
| "mean_token_accuracy": 0.6889510739594698, |
| "num_tokens": 1979232.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.3617051038891077, |
| "epoch": 0.128, |
| "grad_norm": 31.0, |
| "learning_rate": 5.2666666666666665e-06, |
| "loss": 23.138, |
| "mean_token_accuracy": 0.701369734108448, |
| "num_tokens": 2274177.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.3085643179714679, |
| "epoch": 0.144, |
| "grad_norm": 26.375, |
| "learning_rate": 5.933333333333335e-06, |
| "loss": 21.2975, |
| "mean_token_accuracy": 0.7145243223756552, |
| "num_tokens": 2548445.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.2485874235630035, |
| "epoch": 0.16, |
| "grad_norm": 28.25, |
| "learning_rate": 6.600000000000001e-06, |
| "loss": 19.7756, |
| "mean_token_accuracy": 0.7278878930956125, |
| "num_tokens": 2824418.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 3.488466117858887, |
| "eval_biology_loss": 4.677812576293945, |
| "eval_biology_mean_token_accuracy": 0.3829730122089386, |
| "eval_biology_num_tokens": 2824418.0, |
| "eval_biology_runtime": 50.0839, |
| "eval_biology_samples_per_second": 9.983, |
| "eval_biology_steps_per_second": 2.496, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 2.3915954236984254, |
| "eval_chemistry_loss": 2.5027005672454834, |
| "eval_chemistry_mean_token_accuracy": 0.5624839839935303, |
| "eval_chemistry_num_tokens": 2824418.0, |
| "eval_chemistry_runtime": 62.0455, |
| "eval_chemistry_samples_per_second": 8.059, |
| "eval_chemistry_steps_per_second": 2.015, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 1.263984902381897, |
| "eval_math_loss": 1.2374709844589233, |
| "eval_math_mean_token_accuracy": 0.7252511687278748, |
| "eval_math_num_tokens": 2824418.0, |
| "eval_math_runtime": 63.6691, |
| "eval_math_samples_per_second": 7.853, |
| "eval_math_steps_per_second": 1.963, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 1.9593838052749635, |
| "eval_physics_loss": 1.9986088275909424, |
| "eval_physics_mean_token_accuracy": 0.6278775181770325, |
| "eval_physics_num_tokens": 2824418.0, |
| "eval_physics_runtime": 72.5186, |
| "eval_physics_samples_per_second": 6.895, |
| "eval_physics_steps_per_second": 1.724, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.2179695062339306, |
| "epoch": 0.176, |
| "grad_norm": 22.75, |
| "learning_rate": 7.266666666666668e-06, |
| "loss": 19.1845, |
| "mean_token_accuracy": 0.7315629377961159, |
| "num_tokens": 3110313.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.2148031558841468, |
| "epoch": 0.192, |
| "grad_norm": 21.625, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 18.9916, |
| "mean_token_accuracy": 0.7303893141448498, |
| "num_tokens": 3394170.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.1610155675560236, |
| "epoch": 0.208, |
| "grad_norm": 19.25, |
| "learning_rate": 8.6e-06, |
| "loss": 18.5031, |
| "mean_token_accuracy": 0.7372519567608833, |
| "num_tokens": 3673600.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.1351299952715634, |
| "epoch": 0.224, |
| "grad_norm": 21.75, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 18.0553, |
| "mean_token_accuracy": 0.7397045020014048, |
| "num_tokens": 3953732.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.0942034516483545, |
| "epoch": 0.24, |
| "grad_norm": 20.75, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 17.3542, |
| "mean_token_accuracy": 0.7487591315060854, |
| "num_tokens": 4243655.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.0573217056691646, |
| "epoch": 0.256, |
| "grad_norm": 19.75, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 16.8031, |
| "mean_token_accuracy": 0.7541655786335468, |
| "num_tokens": 4531471.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.0622627913951874, |
| "epoch": 0.272, |
| "grad_norm": 17.25, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 16.9748, |
| "mean_token_accuracy": 0.7531674332916737, |
| "num_tokens": 4810284.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.0466016590595246, |
| "epoch": 0.288, |
| "grad_norm": 15.8125, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 16.5926, |
| "mean_token_accuracy": 0.7563348900526762, |
| "num_tokens": 5095104.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.0420017059892417, |
| "epoch": 0.304, |
| "grad_norm": 15.625, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 16.6259, |
| "mean_token_accuracy": 0.7554979339241982, |
| "num_tokens": 5383732.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.0120286911725997, |
| "epoch": 0.32, |
| "grad_norm": 18.25, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 16.1236, |
| "mean_token_accuracy": 0.7615335866808891, |
| "num_tokens": 5676334.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 3.398402997970581, |
| "eval_biology_loss": 4.174577236175537, |
| "eval_biology_mean_token_accuracy": 0.40706401348114013, |
| "eval_biology_num_tokens": 5676334.0, |
| "eval_biology_runtime": 50.0834, |
| "eval_biology_samples_per_second": 9.983, |
| "eval_biology_steps_per_second": 2.496, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 2.1380633115768433, |
| "eval_chemistry_loss": 2.1777119636535645, |
| "eval_chemistry_mean_token_accuracy": 0.5923420660495758, |
| "eval_chemistry_num_tokens": 5676334.0, |
| "eval_chemistry_runtime": 62.077, |
| "eval_chemistry_samples_per_second": 8.055, |
| "eval_chemistry_steps_per_second": 2.014, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 1.0378505063056946, |
| "eval_math_loss": 1.031339168548584, |
| "eval_math_mean_token_accuracy": 0.7560033488273621, |
| "eval_math_num_tokens": 5676334.0, |
| "eval_math_runtime": 63.6444, |
| "eval_math_samples_per_second": 7.856, |
| "eval_math_steps_per_second": 1.964, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 1.6895962080955504, |
| "eval_physics_loss": 1.734020471572876, |
| "eval_physics_mean_token_accuracy": 0.6560393133163452, |
| "eval_physics_num_tokens": 5676334.0, |
| "eval_physics_runtime": 72.4962, |
| "eval_physics_samples_per_second": 6.897, |
| "eval_physics_steps_per_second": 1.724, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.0268741879612207, |
| "epoch": 0.336, |
| "grad_norm": 18.0, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 16.46, |
| "mean_token_accuracy": 0.7554850295186043, |
| "num_tokens": 5958480.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.9965500451624394, |
| "epoch": 0.352, |
| "grad_norm": 16.375, |
| "learning_rate": 1.46e-05, |
| "loss": 15.7948, |
| "mean_token_accuracy": 0.7630108542740345, |
| "num_tokens": 6242161.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.994103953614831, |
| "epoch": 0.368, |
| "grad_norm": 17.875, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 15.8963, |
| "mean_token_accuracy": 0.7624320674687624, |
| "num_tokens": 6523679.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.9746689781546592, |
| "epoch": 0.384, |
| "grad_norm": 16.875, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 15.5515, |
| "mean_token_accuracy": 0.7664125367999077, |
| "num_tokens": 6810978.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.0081916853785515, |
| "epoch": 0.4, |
| "grad_norm": 16.5, |
| "learning_rate": 1.66e-05, |
| "loss": 16.4028, |
| "mean_token_accuracy": 0.7586446344852448, |
| "num_tokens": 7096903.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.9798103597015142, |
| "epoch": 0.416, |
| "grad_norm": 15.75, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 15.6138, |
| "mean_token_accuracy": 0.7639607947319746, |
| "num_tokens": 7377181.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.9465980283915997, |
| "epoch": 0.432, |
| "grad_norm": 17.625, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 15.2106, |
| "mean_token_accuracy": 0.7715261545032263, |
| "num_tokens": 7650523.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.9358791437000036, |
| "epoch": 0.448, |
| "grad_norm": 18.875, |
| "learning_rate": 1.86e-05, |
| "loss": 15.0287, |
| "mean_token_accuracy": 0.771769804134965, |
| "num_tokens": 7936788.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.9421842556446791, |
| "epoch": 0.464, |
| "grad_norm": 20.0, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 15.2138, |
| "mean_token_accuracy": 0.7710198521614074, |
| "num_tokens": 8223147.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.962396053224802, |
| "epoch": 0.48, |
| "grad_norm": 17.75, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 15.3417, |
| "mean_token_accuracy": 0.7693098299205303, |
| "num_tokens": 8506030.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 3.405539764404297, |
| "eval_biology_loss": 3.895674228668213, |
| "eval_biology_mean_token_accuracy": 0.4204931092262268, |
| "eval_biology_num_tokens": 8506030.0, |
| "eval_biology_runtime": 50.1367, |
| "eval_biology_samples_per_second": 9.973, |
| "eval_biology_steps_per_second": 2.493, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 2.0477921390533447, |
| "eval_chemistry_loss": 2.050157308578491, |
| "eval_chemistry_mean_token_accuracy": 0.6057685849666595, |
| "eval_chemistry_num_tokens": 8506030.0, |
| "eval_chemistry_runtime": 62.1697, |
| "eval_chemistry_samples_per_second": 8.043, |
| "eval_chemistry_steps_per_second": 2.011, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 0.9374664115905762, |
| "eval_math_loss": 0.9544177651405334, |
| "eval_math_mean_token_accuracy": 0.7683046078681945, |
| "eval_math_num_tokens": 8506030.0, |
| "eval_math_runtime": 63.7112, |
| "eval_math_samples_per_second": 7.848, |
| "eval_math_steps_per_second": 1.962, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 1.6079197850227356, |
| "eval_physics_loss": 1.6351362466812134, |
| "eval_physics_mean_token_accuracy": 0.6683077902793885, |
| "eval_physics_num_tokens": 8506030.0, |
| "eval_physics_runtime": 72.5398, |
| "eval_physics_samples_per_second": 6.893, |
| "eval_physics_steps_per_second": 1.723, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.9215071354061365, |
| "epoch": 0.496, |
| "grad_norm": 26.875, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 14.792, |
| "mean_token_accuracy": 0.7745936445891857, |
| "num_tokens": 8788726.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.9270129840821028, |
| "epoch": 0.512, |
| "grad_norm": 17.5, |
| "learning_rate": 1.985925925925926e-05, |
| "loss": 14.8253, |
| "mean_token_accuracy": 0.7746361777186394, |
| "num_tokens": 9078039.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.899552209302783, |
| "epoch": 0.528, |
| "grad_norm": 20.125, |
| "learning_rate": 1.9785185185185187e-05, |
| "loss": 14.5531, |
| "mean_token_accuracy": 0.7795046135783196, |
| "num_tokens": 9373860.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.9075620226562023, |
| "epoch": 0.544, |
| "grad_norm": 17.875, |
| "learning_rate": 1.971111111111111e-05, |
| "loss": 14.713, |
| "mean_token_accuracy": 0.777441994100809, |
| "num_tokens": 9660940.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.9506530221551657, |
| "epoch": 0.56, |
| "grad_norm": 25.25, |
| "learning_rate": 1.963703703703704e-05, |
| "loss": 15.1709, |
| "mean_token_accuracy": 0.7681664042174816, |
| "num_tokens": 9932302.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.9100017622113228, |
| "epoch": 0.576, |
| "grad_norm": 18.0, |
| "learning_rate": 1.9562962962962964e-05, |
| "loss": 14.6505, |
| "mean_token_accuracy": 0.7754532121121883, |
| "num_tokens": 10215462.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.896766658872366, |
| "epoch": 0.592, |
| "grad_norm": 16.125, |
| "learning_rate": 1.948888888888889e-05, |
| "loss": 14.5203, |
| "mean_token_accuracy": 0.7769503209739923, |
| "num_tokens": 10504396.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.9094142084941268, |
| "epoch": 0.608, |
| "grad_norm": 16.25, |
| "learning_rate": 1.9414814814814817e-05, |
| "loss": 14.6279, |
| "mean_token_accuracy": 0.7770587045699358, |
| "num_tokens": 10793126.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.8876220740377903, |
| "epoch": 0.624, |
| "grad_norm": 18.0, |
| "learning_rate": 1.9340740740740743e-05, |
| "loss": 14.3233, |
| "mean_token_accuracy": 0.7783829730004073, |
| "num_tokens": 11081768.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.88020947445184, |
| "epoch": 0.64, |
| "grad_norm": 16.75, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 14.0524, |
| "mean_token_accuracy": 0.780827208608389, |
| "num_tokens": 11370320.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 3.335602487564087, |
| "eval_biology_loss": 3.8722915649414062, |
| "eval_biology_mean_token_accuracy": 0.42685190558433533, |
| "eval_biology_num_tokens": 11370320.0, |
| "eval_biology_runtime": 50.0979, |
| "eval_biology_samples_per_second": 9.98, |
| "eval_biology_steps_per_second": 2.495, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 1.9703331971168518, |
| "eval_chemistry_loss": 1.9906623363494873, |
| "eval_chemistry_mean_token_accuracy": 0.6160369579792022, |
| "eval_chemistry_num_tokens": 11370320.0, |
| "eval_chemistry_runtime": 62.0625, |
| "eval_chemistry_samples_per_second": 8.056, |
| "eval_chemistry_steps_per_second": 2.014, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 0.9068713026046753, |
| "eval_math_loss": 0.910418689250946, |
| "eval_math_mean_token_accuracy": 0.7759209518432617, |
| "eval_math_num_tokens": 11370320.0, |
| "eval_math_runtime": 63.6487, |
| "eval_math_samples_per_second": 7.856, |
| "eval_math_steps_per_second": 1.964, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 1.5451811146736145, |
| "eval_physics_loss": 1.5735290050506592, |
| "eval_physics_mean_token_accuracy": 0.6776216115951538, |
| "eval_physics_num_tokens": 11370320.0, |
| "eval_physics_runtime": 72.4949, |
| "eval_physics_samples_per_second": 6.897, |
| "eval_physics_steps_per_second": 1.724, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.860218596458435, |
| "epoch": 0.656, |
| "grad_norm": 15.0625, |
| "learning_rate": 1.9192592592592593e-05, |
| "loss": 13.8789, |
| "mean_token_accuracy": 0.7833294775336981, |
| "num_tokens": 11657835.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.8948563069105149, |
| "epoch": 0.672, |
| "grad_norm": 16.5, |
| "learning_rate": 1.911851851851852e-05, |
| "loss": 14.3792, |
| "mean_token_accuracy": 0.7774154812097549, |
| "num_tokens": 11949262.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.8625882860273123, |
| "epoch": 0.688, |
| "grad_norm": 14.0, |
| "learning_rate": 1.9044444444444446e-05, |
| "loss": 13.8761, |
| "mean_token_accuracy": 0.7837949499487877, |
| "num_tokens": 12227640.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.8663179978728295, |
| "epoch": 0.704, |
| "grad_norm": 16.25, |
| "learning_rate": 1.8970370370370372e-05, |
| "loss": 13.9767, |
| "mean_token_accuracy": 0.7831656277179718, |
| "num_tokens": 12516641.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.8871479570865631, |
| "epoch": 0.72, |
| "grad_norm": 18.125, |
| "learning_rate": 1.8896296296296295e-05, |
| "loss": 14.3515, |
| "mean_token_accuracy": 0.7785210806876421, |
| "num_tokens": 12793343.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.872298126295209, |
| "epoch": 0.736, |
| "grad_norm": 15.875, |
| "learning_rate": 1.8822222222222225e-05, |
| "loss": 14.0105, |
| "mean_token_accuracy": 0.7827250022441149, |
| "num_tokens": 13077981.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.8562807805836201, |
| "epoch": 0.752, |
| "grad_norm": 17.375, |
| "learning_rate": 1.874814814814815e-05, |
| "loss": 13.6996, |
| "mean_token_accuracy": 0.7862995523959398, |
| "num_tokens": 13358957.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.8273469803854823, |
| "epoch": 0.768, |
| "grad_norm": 14.1875, |
| "learning_rate": 1.8674074074074075e-05, |
| "loss": 13.3777, |
| "mean_token_accuracy": 0.7925803154706955, |
| "num_tokens": 13653412.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.8882941573858261, |
| "epoch": 0.784, |
| "grad_norm": 17.25, |
| "learning_rate": 1.86e-05, |
| "loss": 14.3299, |
| "mean_token_accuracy": 0.7802201680839062, |
| "num_tokens": 13940856.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.8893520476296544, |
| "epoch": 0.8, |
| "grad_norm": 17.25, |
| "learning_rate": 1.8525925925925928e-05, |
| "loss": 14.3909, |
| "mean_token_accuracy": 0.7795332457870245, |
| "num_tokens": 14230754.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 3.442336763381958, |
| "eval_biology_loss": 3.7037065029144287, |
| "eval_biology_mean_token_accuracy": 0.43281020307540896, |
| "eval_biology_num_tokens": 14230754.0, |
| "eval_biology_runtime": 49.9381, |
| "eval_biology_samples_per_second": 10.012, |
| "eval_biology_steps_per_second": 2.503, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 1.960342116355896, |
| "eval_chemistry_loss": 1.9245615005493164, |
| "eval_chemistry_mean_token_accuracy": 0.6227234079837799, |
| "eval_chemistry_num_tokens": 14230754.0, |
| "eval_chemistry_runtime": 61.6487, |
| "eval_chemistry_samples_per_second": 8.11, |
| "eval_chemistry_steps_per_second": 2.028, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 0.8902029480934143, |
| "eval_math_loss": 0.885689914226532, |
| "eval_math_mean_token_accuracy": 0.7805563545227051, |
| "eval_math_num_tokens": 14230754.0, |
| "eval_math_runtime": 63.5447, |
| "eval_math_samples_per_second": 7.868, |
| "eval_math_steps_per_second": 1.967, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 1.5454372839927673, |
| "eval_physics_loss": 1.5333291292190552, |
| "eval_physics_mean_token_accuracy": 0.683074339389801, |
| "eval_physics_num_tokens": 14230754.0, |
| "eval_physics_runtime": 72.37, |
| "eval_physics_samples_per_second": 6.909, |
| "eval_physics_steps_per_second": 1.727, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.8682117871940136, |
| "epoch": 0.816, |
| "grad_norm": 16.0, |
| "learning_rate": 1.8451851851851855e-05, |
| "loss": 13.814, |
| "mean_token_accuracy": 0.7841372516006231, |
| "num_tokens": 14519893.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.8578620586544276, |
| "epoch": 0.832, |
| "grad_norm": 16.375, |
| "learning_rate": 1.8377777777777778e-05, |
| "loss": 13.8317, |
| "mean_token_accuracy": 0.7859897501766682, |
| "num_tokens": 14805088.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.8855314686894417, |
| "epoch": 0.848, |
| "grad_norm": 17.125, |
| "learning_rate": 1.8303703703703704e-05, |
| "loss": 14.274, |
| "mean_token_accuracy": 0.7796536423265934, |
| "num_tokens": 15086306.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.8731808762997388, |
| "epoch": 0.864, |
| "grad_norm": 14.1875, |
| "learning_rate": 1.822962962962963e-05, |
| "loss": 14.0156, |
| "mean_token_accuracy": 0.7828668527305126, |
| "num_tokens": 15370985.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.880347841233015, |
| "epoch": 0.88, |
| "grad_norm": 19.375, |
| "learning_rate": 1.8155555555555557e-05, |
| "loss": 14.1502, |
| "mean_token_accuracy": 0.7809113275259734, |
| "num_tokens": 15653836.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.8649558831006289, |
| "epoch": 0.896, |
| "grad_norm": 20.0, |
| "learning_rate": 1.8081481481481484e-05, |
| "loss": 13.9752, |
| "mean_token_accuracy": 0.7830663651227951, |
| "num_tokens": 15932179.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.8504776157438755, |
| "epoch": 0.912, |
| "grad_norm": 16.625, |
| "learning_rate": 1.800740740740741e-05, |
| "loss": 13.5728, |
| "mean_token_accuracy": 0.7864992685616017, |
| "num_tokens": 16219842.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.8390729090198874, |
| "epoch": 0.928, |
| "grad_norm": 19.25, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 13.5587, |
| "mean_token_accuracy": 0.7872791964560747, |
| "num_tokens": 16499842.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.865405573323369, |
| "epoch": 0.944, |
| "grad_norm": 31.375, |
| "learning_rate": 1.785925925925926e-05, |
| "loss": 13.9929, |
| "mean_token_accuracy": 0.7827581968158484, |
| "num_tokens": 16781882.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.8798814173787832, |
| "epoch": 0.96, |
| "grad_norm": 14.6875, |
| "learning_rate": 1.7785185185185186e-05, |
| "loss": 14.0639, |
| "mean_token_accuracy": 0.7827699087560177, |
| "num_tokens": 17067407.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 3.278426109313965, |
| "eval_biology_loss": 3.676957607269287, |
| "eval_biology_mean_token_accuracy": 0.4364860031604767, |
| "eval_biology_num_tokens": 17067407.0, |
| "eval_biology_runtime": 49.9669, |
| "eval_biology_samples_per_second": 10.007, |
| "eval_biology_steps_per_second": 2.502, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_chemistry_entropy": 1.857783148765564, |
| "eval_chemistry_loss": 1.8980321884155273, |
| "eval_chemistry_mean_token_accuracy": 0.6259854953289032, |
| "eval_chemistry_num_tokens": 17067407.0, |
| "eval_chemistry_runtime": 62.1468, |
| "eval_chemistry_samples_per_second": 8.045, |
| "eval_chemistry_steps_per_second": 2.011, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_math_entropy": 0.8647160024642945, |
| "eval_math_loss": 0.868224024772644, |
| "eval_math_mean_token_accuracy": 0.7833898453712463, |
| "eval_math_num_tokens": 17067407.0, |
| "eval_math_runtime": 63.5747, |
| "eval_math_samples_per_second": 7.865, |
| "eval_math_steps_per_second": 1.966, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_physics_entropy": 1.4752862091064454, |
| "eval_physics_loss": 1.513607144355774, |
| "eval_physics_mean_token_accuracy": 0.6863754806518555, |
| "eval_physics_num_tokens": 17067407.0, |
| "eval_physics_runtime": 72.354, |
| "eval_physics_samples_per_second": 6.91, |
| "eval_physics_steps_per_second": 1.728, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.8497918715700508, |
| "epoch": 0.976, |
| "grad_norm": 18.625, |
| "learning_rate": 1.7711111111111113e-05, |
| "loss": 13.6664, |
| "mean_token_accuracy": 0.7879329767078161, |
| "num_tokens": 17350994.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.86589151956141, |
| "epoch": 0.992, |
| "grad_norm": 16.25, |
| "learning_rate": 1.763703703703704e-05, |
| "loss": 13.9208, |
| "mean_token_accuracy": 0.7827646978199482, |
| "num_tokens": 17637514.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.8501669876277447, |
| "epoch": 1.008, |
| "grad_norm": 15.375, |
| "learning_rate": 1.7562962962962962e-05, |
| "loss": 13.5677, |
| "mean_token_accuracy": 0.7862402476370335, |
| "num_tokens": 17926570.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.8345263335853815, |
| "epoch": 1.024, |
| "grad_norm": 15.625, |
| "learning_rate": 1.7488888888888892e-05, |
| "loss": 13.3578, |
| "mean_token_accuracy": 0.7888704732060432, |
| "num_tokens": 18207652.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.8276458965614438, |
| "epoch": 1.04, |
| "grad_norm": 17.5, |
| "learning_rate": 1.7414814814814815e-05, |
| "loss": 13.3452, |
| "mean_token_accuracy": 0.7908481806516647, |
| "num_tokens": 18484931.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.8310236493125558, |
| "epoch": 1.056, |
| "grad_norm": 16.875, |
| "learning_rate": 1.7340740740740742e-05, |
| "loss": 13.5223, |
| "mean_token_accuracy": 0.7887609034776688, |
| "num_tokens": 18773457.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.840316267311573, |
| "epoch": 1.072, |
| "grad_norm": 16.625, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 13.4228, |
| "mean_token_accuracy": 0.7874431516975164, |
| "num_tokens": 19055365.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.8311731787398458, |
| "epoch": 1.088, |
| "grad_norm": 14.4375, |
| "learning_rate": 1.7192592592592595e-05, |
| "loss": 13.3971, |
| "mean_token_accuracy": 0.790545703843236, |
| "num_tokens": 19345730.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.8338565738871694, |
| "epoch": 1.104, |
| "grad_norm": 13.6875, |
| "learning_rate": 1.711851851851852e-05, |
| "loss": 13.39, |
| "mean_token_accuracy": 0.7901448987424373, |
| "num_tokens": 19637390.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.8102976515889168, |
| "epoch": 1.12, |
| "grad_norm": 17.5, |
| "learning_rate": 1.7044444444444445e-05, |
| "loss": 12.9856, |
| "mean_token_accuracy": 0.7944051265716553, |
| "num_tokens": 19923914.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_biology_entropy": 3.2230891466140745, |
| "eval_biology_loss": 3.734968423843384, |
| "eval_biology_mean_token_accuracy": 0.4355302617549896, |
| "eval_biology_num_tokens": 19923914.0, |
| "eval_biology_runtime": 49.8901, |
| "eval_biology_samples_per_second": 10.022, |
| "eval_biology_steps_per_second": 2.506, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_chemistry_entropy": 1.7908948783874512, |
| "eval_chemistry_loss": 1.90095853805542, |
| "eval_chemistry_mean_token_accuracy": 0.6279562847614288, |
| "eval_chemistry_num_tokens": 19923914.0, |
| "eval_chemistry_runtime": 61.9568, |
| "eval_chemistry_samples_per_second": 8.07, |
| "eval_chemistry_steps_per_second": 2.018, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_math_entropy": 0.8332763781547546, |
| "eval_math_loss": 0.8580161333084106, |
| "eval_math_mean_token_accuracy": 0.7852629780769348, |
| "eval_math_num_tokens": 19923914.0, |
| "eval_math_runtime": 63.5133, |
| "eval_math_samples_per_second": 7.872, |
| "eval_math_steps_per_second": 1.968, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_physics_entropy": 1.421106719970703, |
| "eval_physics_loss": 1.504669189453125, |
| "eval_physics_mean_token_accuracy": 0.687330924987793, |
| "eval_physics_num_tokens": 19923914.0, |
| "eval_physics_runtime": 72.2525, |
| "eval_physics_samples_per_second": 6.92, |
| "eval_physics_steps_per_second": 1.73, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.8049196273088455, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 16.125, |
| "learning_rate": 1.697037037037037e-05, |
| "loss": 13.0378, |
| "mean_token_accuracy": 0.7929212123155593, |
| "num_tokens": 20201892.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.8303030036389828, |
| "epoch": 1.152, |
| "grad_norm": 17.75, |
| "learning_rate": 1.6896296296296298e-05, |
| "loss": 13.304, |
| "mean_token_accuracy": 0.7919872298836708, |
| "num_tokens": 20490282.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.8234290273860096, |
| "epoch": 1.168, |
| "grad_norm": 14.6875, |
| "learning_rate": 1.6822222222222224e-05, |
| "loss": 13.2574, |
| "mean_token_accuracy": 0.791878254711628, |
| "num_tokens": 20785786.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.8201817765831947, |
| "epoch": 1.184, |
| "grad_norm": 18.25, |
| "learning_rate": 1.6748148148148147e-05, |
| "loss": 13.1797, |
| "mean_token_accuracy": 0.7926335010677577, |
| "num_tokens": 21074205.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.8103797230869532, |
| "epoch": 1.2, |
| "grad_norm": 17.75, |
| "learning_rate": 1.6674074074074077e-05, |
| "loss": 13.0375, |
| "mean_token_accuracy": 0.7937572699040174, |
| "num_tokens": 21369159.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.8331582929939032, |
| "epoch": 1.216, |
| "grad_norm": 21.5, |
| "learning_rate": 1.66e-05, |
| "loss": 13.3733, |
| "mean_token_accuracy": 0.7907441522926092, |
| "num_tokens": 21649178.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.8037656577304005, |
| "epoch": 1.232, |
| "grad_norm": 17.125, |
| "learning_rate": 1.6525925925925927e-05, |
| "loss": 12.8855, |
| "mean_token_accuracy": 0.7958086896687746, |
| "num_tokens": 21930239.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.7982386674731969, |
| "epoch": 1.248, |
| "grad_norm": 15.625, |
| "learning_rate": 1.6451851851851853e-05, |
| "loss": 12.8508, |
| "mean_token_accuracy": 0.7953327149152756, |
| "num_tokens": 22216387.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.8371293131262064, |
| "epoch": 1.264, |
| "grad_norm": 18.625, |
| "learning_rate": 1.637777777777778e-05, |
| "loss": 13.5088, |
| "mean_token_accuracy": 0.7873477075248957, |
| "num_tokens": 22501002.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.8327024061232805, |
| "epoch": 1.28, |
| "grad_norm": 20.625, |
| "learning_rate": 1.6303703703703706e-05, |
| "loss": 13.3064, |
| "mean_token_accuracy": 0.791947865113616, |
| "num_tokens": 22779682.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_biology_entropy": 3.294544553756714, |
| "eval_biology_loss": 3.6706202030181885, |
| "eval_biology_mean_token_accuracy": 0.43737580394744874, |
| "eval_biology_num_tokens": 22779682.0, |
| "eval_biology_runtime": 49.9678, |
| "eval_biology_samples_per_second": 10.006, |
| "eval_biology_steps_per_second": 2.502, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_chemistry_entropy": 1.8024000110626222, |
| "eval_chemistry_loss": 1.8754929304122925, |
| "eval_chemistry_mean_token_accuracy": 0.6302983772754669, |
| "eval_chemistry_num_tokens": 22779682.0, |
| "eval_chemistry_runtime": 62.133, |
| "eval_chemistry_samples_per_second": 8.047, |
| "eval_chemistry_steps_per_second": 2.012, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_math_entropy": 0.8196169562339782, |
| "eval_math_loss": 0.8479236364364624, |
| "eval_math_mean_token_accuracy": 0.7868858284950256, |
| "eval_math_num_tokens": 22779682.0, |
| "eval_math_runtime": 64.1319, |
| "eval_math_samples_per_second": 7.796, |
| "eval_math_steps_per_second": 1.949, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_physics_entropy": 1.4233626689910888, |
| "eval_physics_loss": 1.4852280616760254, |
| "eval_physics_mean_token_accuracy": 0.6896569218635559, |
| "eval_physics_num_tokens": 22779682.0, |
| "eval_physics_runtime": 72.356, |
| "eval_physics_samples_per_second": 6.91, |
| "eval_physics_steps_per_second": 1.728, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.8020031837746501, |
| "epoch": 1.296, |
| "grad_norm": 16.25, |
| "learning_rate": 1.622962962962963e-05, |
| "loss": 12.998, |
| "mean_token_accuracy": 0.7940779849886894, |
| "num_tokens": 23057744.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.812518722563982, |
| "epoch": 1.312, |
| "grad_norm": 19.125, |
| "learning_rate": 1.6155555555555556e-05, |
| "loss": 13.2282, |
| "mean_token_accuracy": 0.7920650199055672, |
| "num_tokens": 23344644.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.8391162436455488, |
| "epoch": 1.328, |
| "grad_norm": 17.875, |
| "learning_rate": 1.6081481481481482e-05, |
| "loss": 13.3581, |
| "mean_token_accuracy": 0.7894898921251297, |
| "num_tokens": 23622405.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.8036098016425968, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 15.6875, |
| "learning_rate": 1.600740740740741e-05, |
| "loss": 12.9155, |
| "mean_token_accuracy": 0.7945138659328222, |
| "num_tokens": 23899771.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.802362441085279, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 14.5, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 12.9723, |
| "mean_token_accuracy": 0.7945049315690994, |
| "num_tokens": 24187023.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.8688682168722153, |
| "epoch": 1.376, |
| "grad_norm": 19.75, |
| "learning_rate": 1.5859259259259262e-05, |
| "loss": 13.9796, |
| "mean_token_accuracy": 0.7812775958329439, |
| "num_tokens": 24466132.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.8100431380793452, |
| "epoch": 1.392, |
| "grad_norm": 18.75, |
| "learning_rate": 1.5785185185185185e-05, |
| "loss": 13.0219, |
| "mean_token_accuracy": 0.7929740186780692, |
| "num_tokens": 24748043.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.7973114985972642, |
| "epoch": 1.408, |
| "grad_norm": 18.5, |
| "learning_rate": 1.571111111111111e-05, |
| "loss": 12.8964, |
| "mean_token_accuracy": 0.7961425919085741, |
| "num_tokens": 25036674.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.7954017139971257, |
| "epoch": 1.424, |
| "grad_norm": 13.6875, |
| "learning_rate": 1.5637037037037038e-05, |
| "loss": 12.8143, |
| "mean_token_accuracy": 0.7969338528811931, |
| "num_tokens": 25324579.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.8209576582536101, |
| "epoch": 1.44, |
| "grad_norm": 18.0, |
| "learning_rate": 1.5562962962962965e-05, |
| "loss": 13.2807, |
| "mean_token_accuracy": 0.7910069234669208, |
| "num_tokens": 25606824.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_biology_entropy": 3.271074831008911, |
| "eval_biology_loss": 3.627021074295044, |
| "eval_biology_mean_token_accuracy": 0.43953498554229736, |
| "eval_biology_num_tokens": 25606824.0, |
| "eval_biology_runtime": 49.8927, |
| "eval_biology_samples_per_second": 10.021, |
| "eval_biology_steps_per_second": 2.505, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_chemistry_entropy": 1.8041245365142822, |
| "eval_chemistry_loss": 1.8709218502044678, |
| "eval_chemistry_mean_token_accuracy": 0.6310126931667328, |
| "eval_chemistry_num_tokens": 25606824.0, |
| "eval_chemistry_runtime": 62.0005, |
| "eval_chemistry_samples_per_second": 8.064, |
| "eval_chemistry_steps_per_second": 2.016, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_math_entropy": 0.8287842729091645, |
| "eval_math_loss": 0.8401960134506226, |
| "eval_math_mean_token_accuracy": 0.7886354041099548, |
| "eval_math_num_tokens": 25606824.0, |
| "eval_math_runtime": 63.5552, |
| "eval_math_samples_per_second": 7.867, |
| "eval_math_steps_per_second": 1.967, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_physics_entropy": 1.4311670818328857, |
| "eval_physics_loss": 1.4744281768798828, |
| "eval_physics_mean_token_accuracy": 0.6918323221206665, |
| "eval_physics_num_tokens": 25606824.0, |
| "eval_physics_runtime": 72.5619, |
| "eval_physics_samples_per_second": 6.891, |
| "eval_physics_steps_per_second": 1.723, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.8060221856459975, |
| "epoch": 1.456, |
| "grad_norm": 14.8125, |
| "learning_rate": 1.548888888888889e-05, |
| "loss": 12.8502, |
| "mean_token_accuracy": 0.7942074920982123, |
| "num_tokens": 25886396.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.804918241687119, |
| "epoch": 1.472, |
| "grad_norm": 16.5, |
| "learning_rate": 1.5414814814814814e-05, |
| "loss": 12.9431, |
| "mean_token_accuracy": 0.794098413363099, |
| "num_tokens": 26163618.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.8308094689622522, |
| "epoch": 1.488, |
| "grad_norm": 17.375, |
| "learning_rate": 1.5340740740740744e-05, |
| "loss": 13.5052, |
| "mean_token_accuracy": 0.7875379234552383, |
| "num_tokens": 26438338.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.7895203305408358, |
| "epoch": 1.504, |
| "grad_norm": 20.25, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 12.7945, |
| "mean_token_accuracy": 0.796988645941019, |
| "num_tokens": 26729255.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.8357090651988983, |
| "epoch": 1.52, |
| "grad_norm": 15.5, |
| "learning_rate": 1.5192592592592594e-05, |
| "loss": 13.3142, |
| "mean_token_accuracy": 0.790850094333291, |
| "num_tokens": 27017935.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.8033582132309676, |
| "epoch": 1.536, |
| "grad_norm": 14.8125, |
| "learning_rate": 1.5118518518518519e-05, |
| "loss": 12.9716, |
| "mean_token_accuracy": 0.7936928037554025, |
| "num_tokens": 27306339.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.8104204999282956, |
| "epoch": 1.552, |
| "grad_norm": 16.625, |
| "learning_rate": 1.5044444444444445e-05, |
| "loss": 12.981, |
| "mean_token_accuracy": 0.7937849141657353, |
| "num_tokens": 27591959.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.7846602015197277, |
| "epoch": 1.568, |
| "grad_norm": 14.0625, |
| "learning_rate": 1.497037037037037e-05, |
| "loss": 12.677, |
| "mean_token_accuracy": 0.7991001665592193, |
| "num_tokens": 27884398.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.7992334404960275, |
| "epoch": 1.584, |
| "grad_norm": 18.875, |
| "learning_rate": 1.4896296296296298e-05, |
| "loss": 12.935, |
| "mean_token_accuracy": 0.794677285850048, |
| "num_tokens": 28171274.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.8037812670692801, |
| "epoch": 1.6, |
| "grad_norm": 19.5, |
| "learning_rate": 1.4822222222222225e-05, |
| "loss": 12.7702, |
| "mean_token_accuracy": 0.7950391560792923, |
| "num_tokens": 28457624.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_biology_entropy": 3.2400936613082885, |
| "eval_biology_loss": 3.704118251800537, |
| "eval_biology_mean_token_accuracy": 0.43956737661361694, |
| "eval_biology_num_tokens": 28457624.0, |
| "eval_biology_runtime": 49.8865, |
| "eval_biology_samples_per_second": 10.023, |
| "eval_biology_steps_per_second": 2.506, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_chemistry_entropy": 1.776292507648468, |
| "eval_chemistry_loss": 1.871563196182251, |
| "eval_chemistry_mean_token_accuracy": 0.632093154668808, |
| "eval_chemistry_num_tokens": 28457624.0, |
| "eval_chemistry_runtime": 61.9067, |
| "eval_chemistry_samples_per_second": 8.077, |
| "eval_chemistry_steps_per_second": 2.019, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_math_entropy": 0.8142018506526947, |
| "eval_math_loss": 0.834656834602356, |
| "eval_math_mean_token_accuracy": 0.7897459998130798, |
| "eval_math_num_tokens": 28457624.0, |
| "eval_math_runtime": 64.1472, |
| "eval_math_samples_per_second": 7.795, |
| "eval_math_steps_per_second": 1.949, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_physics_entropy": 1.403171688079834, |
| "eval_physics_loss": 1.471365213394165, |
| "eval_physics_mean_token_accuracy": 0.692293493270874, |
| "eval_physics_num_tokens": 28457624.0, |
| "eval_physics_runtime": 72.7239, |
| "eval_physics_samples_per_second": 6.875, |
| "eval_physics_steps_per_second": 1.719, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.798175222799182, |
| "epoch": 1.616, |
| "grad_norm": 15.0, |
| "learning_rate": 1.474814814814815e-05, |
| "loss": 12.9819, |
| "mean_token_accuracy": 0.7934872336685658, |
| "num_tokens": 28743099.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.8180070934817195, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 16.5, |
| "learning_rate": 1.4674074074074076e-05, |
| "loss": 13.0308, |
| "mean_token_accuracy": 0.7923920370638371, |
| "num_tokens": 29017297.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.7903204390779137, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 17.5, |
| "learning_rate": 1.46e-05, |
| "loss": 12.7676, |
| "mean_token_accuracy": 0.7979931388050318, |
| "num_tokens": 29303707.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.7956463107839227, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 16.5, |
| "learning_rate": 1.4525925925925927e-05, |
| "loss": 12.793, |
| "mean_token_accuracy": 0.7947848867624998, |
| "num_tokens": 29597156.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.8131252270191908, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.4451851851851852e-05, |
| "loss": 13.0555, |
| "mean_token_accuracy": 0.7951273001730442, |
| "num_tokens": 29883879.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.8136409839615226, |
| "epoch": 1.696, |
| "grad_norm": 17.625, |
| "learning_rate": 1.4377777777777779e-05, |
| "loss": 13.173, |
| "mean_token_accuracy": 0.7925207346677781, |
| "num_tokens": 30165760.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.7853763310238719, |
| "epoch": 1.712, |
| "grad_norm": 16.125, |
| "learning_rate": 1.4303703703703703e-05, |
| "loss": 12.7266, |
| "mean_token_accuracy": 0.7975193675607443, |
| "num_tokens": 30460367.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.7903000565245748, |
| "epoch": 1.728, |
| "grad_norm": 21.625, |
| "learning_rate": 1.4229629629629632e-05, |
| "loss": 12.6329, |
| "mean_token_accuracy": 0.7987389255315065, |
| "num_tokens": 30739137.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.8021053321659565, |
| "epoch": 1.744, |
| "grad_norm": 17.625, |
| "learning_rate": 1.4155555555555556e-05, |
| "loss": 12.834, |
| "mean_token_accuracy": 0.7961107190698385, |
| "num_tokens": 31022663.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.7864446669816971, |
| "epoch": 1.76, |
| "grad_norm": 16.625, |
| "learning_rate": 1.4081481481481483e-05, |
| "loss": 12.6971, |
| "mean_token_accuracy": 0.7962396703660488, |
| "num_tokens": 31306494.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_biology_entropy": 3.232375309944153, |
| "eval_biology_loss": 3.648139238357544, |
| "eval_biology_mean_token_accuracy": 0.4412148265838623, |
| "eval_biology_num_tokens": 31306494.0, |
| "eval_biology_runtime": 49.9213, |
| "eval_biology_samples_per_second": 10.016, |
| "eval_biology_steps_per_second": 2.504, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_chemistry_entropy": 1.762632459640503, |
| "eval_chemistry_loss": 1.86100435256958, |
| "eval_chemistry_mean_token_accuracy": 0.6327439775466919, |
| "eval_chemistry_num_tokens": 31306494.0, |
| "eval_chemistry_runtime": 61.921, |
| "eval_chemistry_samples_per_second": 8.075, |
| "eval_chemistry_steps_per_second": 2.019, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_math_entropy": 0.8122558786869049, |
| "eval_math_loss": 0.8297922015190125, |
| "eval_math_mean_token_accuracy": 0.7903609256744385, |
| "eval_math_num_tokens": 31306494.0, |
| "eval_math_runtime": 63.4704, |
| "eval_math_samples_per_second": 7.878, |
| "eval_math_steps_per_second": 1.969, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_physics_entropy": 1.4011106457710265, |
| "eval_physics_loss": 1.4673256874084473, |
| "eval_physics_mean_token_accuracy": 0.692846088886261, |
| "eval_physics_num_tokens": 31306494.0, |
| "eval_physics_runtime": 72.2185, |
| "eval_physics_samples_per_second": 6.923, |
| "eval_physics_steps_per_second": 1.731, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.8044054295867682, |
| "epoch": 1.776, |
| "grad_norm": 17.125, |
| "learning_rate": 1.400740740740741e-05, |
| "loss": 12.9493, |
| "mean_token_accuracy": 0.7960755385458469, |
| "num_tokens": 31595542.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.8096846660599113, |
| "epoch": 1.792, |
| "grad_norm": 16.125, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 13.0798, |
| "mean_token_accuracy": 0.7921652976423502, |
| "num_tokens": 31881189.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.8158628990873694, |
| "epoch": 1.808, |
| "grad_norm": 15.1875, |
| "learning_rate": 1.385925925925926e-05, |
| "loss": 12.9956, |
| "mean_token_accuracy": 0.7941187128424645, |
| "num_tokens": 32164196.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.7536683512851596, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 15.75, |
| "learning_rate": 1.3785185185185186e-05, |
| "loss": 12.2054, |
| "mean_token_accuracy": 0.8031131632626056, |
| "num_tokens": 32441530.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.7897313551977276, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 18.625, |
| "learning_rate": 1.3711111111111112e-05, |
| "loss": 12.7218, |
| "mean_token_accuracy": 0.7971313249319791, |
| "num_tokens": 32723145.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.8171826468780636, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 15.125, |
| "learning_rate": 1.3637037037037037e-05, |
| "loss": 13.0642, |
| "mean_token_accuracy": 0.7934288665652275, |
| "num_tokens": 33011263.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.7859557889401912, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 15.75, |
| "learning_rate": 1.3562962962962965e-05, |
| "loss": 12.6724, |
| "mean_token_accuracy": 0.8001265831291675, |
| "num_tokens": 33298921.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.8064542135223747, |
| "epoch": 1.888, |
| "grad_norm": 18.125, |
| "learning_rate": 1.3488888888888888e-05, |
| "loss": 12.9109, |
| "mean_token_accuracy": 0.7947272609919309, |
| "num_tokens": 33576243.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.7957150738686323, |
| "epoch": 1.904, |
| "grad_norm": 15.375, |
| "learning_rate": 1.3414814814814817e-05, |
| "loss": 12.8458, |
| "mean_token_accuracy": 0.7957113239914179, |
| "num_tokens": 33850968.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.7726223962381482, |
| "epoch": 1.92, |
| "grad_norm": 15.125, |
| "learning_rate": 1.3340740740740741e-05, |
| "loss": 12.4511, |
| "mean_token_accuracy": 0.8002966489642859, |
| "num_tokens": 34128558.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_biology_entropy": 3.183071039199829, |
| "eval_biology_loss": 3.584599494934082, |
| "eval_biology_mean_token_accuracy": 0.44353048944473267, |
| "eval_biology_num_tokens": 34128558.0, |
| "eval_biology_runtime": 50.2602, |
| "eval_biology_samples_per_second": 9.948, |
| "eval_biology_steps_per_second": 2.487, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_chemistry_entropy": 1.742585494041443, |
| "eval_chemistry_loss": 1.850415825843811, |
| "eval_chemistry_mean_token_accuracy": 0.6337505750656128, |
| "eval_chemistry_num_tokens": 34128558.0, |
| "eval_chemistry_runtime": 61.9178, |
| "eval_chemistry_samples_per_second": 8.075, |
| "eval_chemistry_steps_per_second": 2.019, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_math_entropy": 0.812834040403366, |
| "eval_math_loss": 0.8256640434265137, |
| "eval_math_mean_token_accuracy": 0.7911388430595397, |
| "eval_math_num_tokens": 34128558.0, |
| "eval_math_runtime": 63.4908, |
| "eval_math_samples_per_second": 7.875, |
| "eval_math_steps_per_second": 1.969, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_physics_entropy": 1.396484270095825, |
| "eval_physics_loss": 1.4663584232330322, |
| "eval_physics_mean_token_accuracy": 0.6928882389068604, |
| "eval_physics_num_tokens": 34128558.0, |
| "eval_physics_runtime": 72.4413, |
| "eval_physics_samples_per_second": 6.902, |
| "eval_physics_steps_per_second": 1.726, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.8009239070117473, |
| "epoch": 1.936, |
| "grad_norm": 21.125, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 12.8074, |
| "mean_token_accuracy": 0.7954005278646946, |
| "num_tokens": 34408056.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.7883490238338708, |
| "epoch": 1.952, |
| "grad_norm": 18.75, |
| "learning_rate": 1.3192592592592594e-05, |
| "loss": 12.7364, |
| "mean_token_accuracy": 0.7968561142683029, |
| "num_tokens": 34684679.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.8134514577686787, |
| "epoch": 1.968, |
| "grad_norm": 17.625, |
| "learning_rate": 1.311851851851852e-05, |
| "loss": 13.2068, |
| "mean_token_accuracy": 0.7917527597397566, |
| "num_tokens": 34971038.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.7896538088098168, |
| "epoch": 1.984, |
| "grad_norm": 16.0, |
| "learning_rate": 1.3044444444444446e-05, |
| "loss": 12.6304, |
| "mean_token_accuracy": 0.7988350588828326, |
| "num_tokens": 35262281.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.7750496102496982, |
| "epoch": 2.0, |
| "grad_norm": 19.25, |
| "learning_rate": 1.297037037037037e-05, |
| "loss": 12.492, |
| "mean_token_accuracy": 0.799928730726242, |
| "num_tokens": 35560864.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.8006911057978868, |
| "epoch": 2.016, |
| "grad_norm": 14.1875, |
| "learning_rate": 1.2896296296296299e-05, |
| "loss": 12.8078, |
| "mean_token_accuracy": 0.7973089322447777, |
| "num_tokens": 35846704.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.7901067767292261, |
| "epoch": 2.032, |
| "grad_norm": 17.625, |
| "learning_rate": 1.2822222222222222e-05, |
| "loss": 12.6518, |
| "mean_token_accuracy": 0.7975228931754827, |
| "num_tokens": 36128775.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.7684864716604352, |
| "epoch": 2.048, |
| "grad_norm": 16.875, |
| "learning_rate": 1.274814814814815e-05, |
| "loss": 12.2821, |
| "mean_token_accuracy": 0.803328375890851, |
| "num_tokens": 36419504.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.7612387906759978, |
| "epoch": 2.064, |
| "grad_norm": 16.25, |
| "learning_rate": 1.2674074074074075e-05, |
| "loss": 12.3434, |
| "mean_token_accuracy": 0.8015977006405592, |
| "num_tokens": 36706816.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.7767295489087701, |
| "epoch": 2.08, |
| "grad_norm": 16.375, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 12.345, |
| "mean_token_accuracy": 0.8008607290685177, |
| "num_tokens": 36988475.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_biology_entropy": 3.112719190597534, |
| "eval_biology_loss": 3.577775716781616, |
| "eval_biology_mean_token_accuracy": 0.4433496446609497, |
| "eval_biology_num_tokens": 36988475.0, |
| "eval_biology_runtime": 50.4638, |
| "eval_biology_samples_per_second": 9.908, |
| "eval_biology_steps_per_second": 2.477, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_chemistry_entropy": 1.6892164297103882, |
| "eval_chemistry_loss": 1.852262258529663, |
| "eval_chemistry_mean_token_accuracy": 0.6334386568069458, |
| "eval_chemistry_num_tokens": 36988475.0, |
| "eval_chemistry_runtime": 62.4497, |
| "eval_chemistry_samples_per_second": 8.006, |
| "eval_chemistry_steps_per_second": 2.002, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_math_entropy": 0.7883058032989502, |
| "eval_math_loss": 0.8236408829689026, |
| "eval_math_mean_token_accuracy": 0.7912847075462341, |
| "eval_math_num_tokens": 36988475.0, |
| "eval_math_runtime": 63.9109, |
| "eval_math_samples_per_second": 7.823, |
| "eval_math_steps_per_second": 1.956, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_physics_entropy": 1.3466296257972716, |
| "eval_physics_loss": 1.459655523300171, |
| "eval_physics_mean_token_accuracy": 0.6937613749504089, |
| "eval_physics_num_tokens": 36988475.0, |
| "eval_physics_runtime": 72.3423, |
| "eval_physics_samples_per_second": 6.912, |
| "eval_physics_steps_per_second": 1.728, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.7833418918773531, |
| "epoch": 2.096, |
| "grad_norm": 15.9375, |
| "learning_rate": 1.2525925925925928e-05, |
| "loss": 12.6724, |
| "mean_token_accuracy": 0.7996709801256656, |
| "num_tokens": 37270131.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.7611293364316225, |
| "epoch": 2.112, |
| "grad_norm": 15.5, |
| "learning_rate": 1.2451851851851853e-05, |
| "loss": 12.2516, |
| "mean_token_accuracy": 0.8012572944164276, |
| "num_tokens": 37563537.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.7628815559670329, |
| "epoch": 2.128, |
| "grad_norm": 18.375, |
| "learning_rate": 1.237777777777778e-05, |
| "loss": 12.157, |
| "mean_token_accuracy": 0.8029138047248126, |
| "num_tokens": 37843959.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.7627110643312335, |
| "epoch": 2.144, |
| "grad_norm": 16.875, |
| "learning_rate": 1.2303703703703704e-05, |
| "loss": 12.3203, |
| "mean_token_accuracy": 0.8017950035631657, |
| "num_tokens": 38133092.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.7732076404616237, |
| "epoch": 2.16, |
| "grad_norm": 15.5625, |
| "learning_rate": 1.222962962962963e-05, |
| "loss": 12.3963, |
| "mean_token_accuracy": 0.8020700648427009, |
| "num_tokens": 38421229.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.7416583737358451, |
| "epoch": 2.176, |
| "grad_norm": 19.375, |
| "learning_rate": 1.2155555555555555e-05, |
| "loss": 11.9055, |
| "mean_token_accuracy": 0.8070738401263953, |
| "num_tokens": 38708043.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.7689750671386719, |
| "epoch": 2.192, |
| "grad_norm": 15.9375, |
| "learning_rate": 1.2081481481481484e-05, |
| "loss": 12.3704, |
| "mean_token_accuracy": 0.800269627571106, |
| "num_tokens": 38996930.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.7766441306099295, |
| "epoch": 2.208, |
| "grad_norm": 29.375, |
| "learning_rate": 1.2007407407407408e-05, |
| "loss": 12.5879, |
| "mean_token_accuracy": 0.7995092123746872, |
| "num_tokens": 39279481.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.7676854202523827, |
| "epoch": 2.224, |
| "grad_norm": 18.375, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 12.3774, |
| "mean_token_accuracy": 0.8020657405257225, |
| "num_tokens": 39569030.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.7721736744046211, |
| "epoch": 2.24, |
| "grad_norm": 16.5, |
| "learning_rate": 1.185925925925926e-05, |
| "loss": 12.385, |
| "mean_token_accuracy": 0.8002489194273948, |
| "num_tokens": 39854873.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_biology_entropy": 3.1739475746154784, |
| "eval_biology_loss": 3.621983528137207, |
| "eval_biology_mean_token_accuracy": 0.4429086031913757, |
| "eval_biology_num_tokens": 39854873.0, |
| "eval_biology_runtime": 50.3472, |
| "eval_biology_samples_per_second": 9.931, |
| "eval_biology_steps_per_second": 2.483, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_chemistry_entropy": 1.7147236638069152, |
| "eval_chemistry_loss": 1.8550182580947876, |
| "eval_chemistry_mean_token_accuracy": 0.6335213797092438, |
| "eval_chemistry_num_tokens": 39854873.0, |
| "eval_chemistry_runtime": 61.9675, |
| "eval_chemistry_samples_per_second": 8.069, |
| "eval_chemistry_steps_per_second": 2.017, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_math_entropy": 0.7918919861316681, |
| "eval_math_loss": 0.8215861320495605, |
| "eval_math_mean_token_accuracy": 0.7919570498466492, |
| "eval_math_num_tokens": 39854873.0, |
| "eval_math_runtime": 63.4845, |
| "eval_math_samples_per_second": 7.876, |
| "eval_math_steps_per_second": 1.969, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_physics_entropy": 1.364860999584198, |
| "eval_physics_loss": 1.4583380222320557, |
| "eval_physics_mean_token_accuracy": 0.6940155153274536, |
| "eval_physics_num_tokens": 39854873.0, |
| "eval_physics_runtime": 72.3589, |
| "eval_physics_samples_per_second": 6.91, |
| "eval_physics_steps_per_second": 1.728, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.7666558466851712, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 16.875, |
| "learning_rate": 1.1785185185185186e-05, |
| "loss": 12.3089, |
| "mean_token_accuracy": 0.8012977659702301, |
| "num_tokens": 40141190.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.7879726745188236, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 20.625, |
| "learning_rate": 1.1711111111111113e-05, |
| "loss": 12.6865, |
| "mean_token_accuracy": 0.798141922056675, |
| "num_tokens": 40415203.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.7712487172335386, |
| "epoch": 2.288, |
| "grad_norm": 16.875, |
| "learning_rate": 1.1637037037037037e-05, |
| "loss": 12.3102, |
| "mean_token_accuracy": 0.800593600794673, |
| "num_tokens": 40702393.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.7664103271439672, |
| "epoch": 2.304, |
| "grad_norm": 20.125, |
| "learning_rate": 1.1562962962962964e-05, |
| "loss": 12.3249, |
| "mean_token_accuracy": 0.8032374054193496, |
| "num_tokens": 40982775.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.7812536470592022, |
| "epoch": 2.32, |
| "grad_norm": 19.375, |
| "learning_rate": 1.1488888888888889e-05, |
| "loss": 12.5473, |
| "mean_token_accuracy": 0.8004047337919473, |
| "num_tokens": 41263356.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.7787353828549385, |
| "epoch": 2.336, |
| "grad_norm": 16.875, |
| "learning_rate": 1.1414814814814817e-05, |
| "loss": 12.5658, |
| "mean_token_accuracy": 0.7996404368430376, |
| "num_tokens": 41545235.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.7861184494569897, |
| "epoch": 2.352, |
| "grad_norm": 17.5, |
| "learning_rate": 1.1340740740740742e-05, |
| "loss": 12.6853, |
| "mean_token_accuracy": 0.7966604385524988, |
| "num_tokens": 41833417.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.772636204212904, |
| "epoch": 2.368, |
| "grad_norm": 18.5, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 12.415, |
| "mean_token_accuracy": 0.8015665214508771, |
| "num_tokens": 42117030.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.7902130583301187, |
| "epoch": 2.384, |
| "grad_norm": 20.375, |
| "learning_rate": 1.1192592592592593e-05, |
| "loss": 12.7256, |
| "mean_token_accuracy": 0.7966745045036078, |
| "num_tokens": 42410990.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.7819203210994601, |
| "epoch": 2.4, |
| "grad_norm": 18.5, |
| "learning_rate": 1.111851851851852e-05, |
| "loss": 12.5447, |
| "mean_token_accuracy": 0.8000232726335526, |
| "num_tokens": 42691890.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_biology_entropy": 3.1204140815734864, |
| "eval_biology_loss": 3.581092357635498, |
| "eval_biology_mean_token_accuracy": 0.4439757652282715, |
| "eval_biology_num_tokens": 42691890.0, |
| "eval_biology_runtime": 50.3432, |
| "eval_biology_samples_per_second": 9.932, |
| "eval_biology_steps_per_second": 2.483, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_chemistry_entropy": 1.677762170791626, |
| "eval_chemistry_loss": 1.8495266437530518, |
| "eval_chemistry_mean_token_accuracy": 0.6339675810337067, |
| "eval_chemistry_num_tokens": 42691890.0, |
| "eval_chemistry_runtime": 61.8947, |
| "eval_chemistry_samples_per_second": 8.078, |
| "eval_chemistry_steps_per_second": 2.02, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_math_entropy": 0.7767732746601105, |
| "eval_math_loss": 0.8190872073173523, |
| "eval_math_mean_token_accuracy": 0.7922054343223571, |
| "eval_math_num_tokens": 42691890.0, |
| "eval_math_runtime": 63.4519, |
| "eval_math_samples_per_second": 7.88, |
| "eval_math_steps_per_second": 1.97, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_physics_entropy": 1.3377611227035522, |
| "eval_physics_loss": 1.4610776901245117, |
| "eval_physics_mean_token_accuracy": 0.6939568905830383, |
| "eval_physics_num_tokens": 42691890.0, |
| "eval_physics_runtime": 72.3367, |
| "eval_physics_samples_per_second": 6.912, |
| "eval_physics_steps_per_second": 1.728, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.7888345830142498, |
| "epoch": 2.416, |
| "grad_norm": 19.5, |
| "learning_rate": 1.1044444444444444e-05, |
| "loss": 12.7716, |
| "mean_token_accuracy": 0.7957693602889776, |
| "num_tokens": 42971588.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.7725735552608967, |
| "epoch": 2.432, |
| "grad_norm": 18.625, |
| "learning_rate": 1.0970370370370371e-05, |
| "loss": 12.355, |
| "mean_token_accuracy": 0.801625344157219, |
| "num_tokens": 43253821.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.7471539022400975, |
| "epoch": 2.448, |
| "grad_norm": 17.625, |
| "learning_rate": 1.0896296296296298e-05, |
| "loss": 12.1773, |
| "mean_token_accuracy": 0.8047775257378816, |
| "num_tokens": 43550902.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.7870838133618235, |
| "epoch": 2.464, |
| "grad_norm": 17.25, |
| "learning_rate": 1.0822222222222222e-05, |
| "loss": 12.4923, |
| "mean_token_accuracy": 0.7991364441812039, |
| "num_tokens": 43844259.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.8207851549610495, |
| "epoch": 2.48, |
| "grad_norm": 25.625, |
| "learning_rate": 1.074814814814815e-05, |
| "loss": 13.1668, |
| "mean_token_accuracy": 0.791607553511858, |
| "num_tokens": 44115701.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.7870116716250777, |
| "epoch": 2.496, |
| "grad_norm": 16.375, |
| "learning_rate": 1.0674074074074074e-05, |
| "loss": 12.7347, |
| "mean_token_accuracy": 0.7973819278180599, |
| "num_tokens": 44405520.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.7786808660253882, |
| "epoch": 2.512, |
| "grad_norm": 15.9375, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 12.4835, |
| "mean_token_accuracy": 0.7992223002016544, |
| "num_tokens": 44686477.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.7790904261171818, |
| "epoch": 2.528, |
| "grad_norm": 16.375, |
| "learning_rate": 1.0525925925925927e-05, |
| "loss": 12.5262, |
| "mean_token_accuracy": 0.7992076646536589, |
| "num_tokens": 44969760.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.771877340041101, |
| "epoch": 2.544, |
| "grad_norm": 17.75, |
| "learning_rate": 1.0451851851851853e-05, |
| "loss": 12.4079, |
| "mean_token_accuracy": 0.8014226417988539, |
| "num_tokens": 45255326.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.7738017693161965, |
| "epoch": 2.56, |
| "grad_norm": 16.125, |
| "learning_rate": 1.0377777777777778e-05, |
| "loss": 12.4005, |
| "mean_token_accuracy": 0.7992622271180153, |
| "num_tokens": 45532525.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_biology_entropy": 3.1479014921188355, |
| "eval_biology_loss": 3.573671340942383, |
| "eval_biology_mean_token_accuracy": 0.44443820452690125, |
| "eval_biology_num_tokens": 45532525.0, |
| "eval_biology_runtime": 49.9183, |
| "eval_biology_samples_per_second": 10.016, |
| "eval_biology_steps_per_second": 2.504, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_chemistry_entropy": 1.6901906566619873, |
| "eval_chemistry_loss": 1.847903847694397, |
| "eval_chemistry_mean_token_accuracy": 0.6340129613876343, |
| "eval_chemistry_num_tokens": 45532525.0, |
| "eval_chemistry_runtime": 62.2403, |
| "eval_chemistry_samples_per_second": 8.033, |
| "eval_chemistry_steps_per_second": 2.008, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_math_entropy": 0.7786712296009064, |
| "eval_math_loss": 0.8165695071220398, |
| "eval_math_mean_token_accuracy": 0.793084463596344, |
| "eval_math_num_tokens": 45532525.0, |
| "eval_math_runtime": 63.4967, |
| "eval_math_samples_per_second": 7.874, |
| "eval_math_steps_per_second": 1.969, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_physics_entropy": 1.3460909366607665, |
| "eval_physics_loss": 1.4585051536560059, |
| "eval_physics_mean_token_accuracy": 0.6942276492118835, |
| "eval_physics_num_tokens": 45532525.0, |
| "eval_physics_runtime": 72.3094, |
| "eval_physics_samples_per_second": 6.915, |
| "eval_physics_steps_per_second": 1.729, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.7676852583885193, |
| "epoch": 2.576, |
| "grad_norm": 17.125, |
| "learning_rate": 1.0303703703703705e-05, |
| "loss": 12.4378, |
| "mean_token_accuracy": 0.8008745819330215, |
| "num_tokens": 45817478.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.7641036305576563, |
| "epoch": 2.592, |
| "grad_norm": 32.75, |
| "learning_rate": 1.0229629629629631e-05, |
| "loss": 12.2379, |
| "mean_token_accuracy": 0.8022371467202902, |
| "num_tokens": 46109575.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.7862739590927958, |
| "epoch": 2.608, |
| "grad_norm": 21.0, |
| "learning_rate": 1.0155555555555556e-05, |
| "loss": 12.7173, |
| "mean_token_accuracy": 0.7964495204389095, |
| "num_tokens": 46391461.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.7508702775463462, |
| "epoch": 2.624, |
| "grad_norm": 16.625, |
| "learning_rate": 1.0081481481481484e-05, |
| "loss": 12.0977, |
| "mean_token_accuracy": 0.8073150396347046, |
| "num_tokens": 46683117.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.7849213285371661, |
| "epoch": 2.64, |
| "grad_norm": 17.5, |
| "learning_rate": 1.0007407407407407e-05, |
| "loss": 12.5349, |
| "mean_token_accuracy": 0.7980777274817228, |
| "num_tokens": 46965534.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.7460307456552983, |
| "epoch": 2.656, |
| "grad_norm": 17.375, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 12.0031, |
| "mean_token_accuracy": 0.8042830608785152, |
| "num_tokens": 47245852.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.7861375520005822, |
| "epoch": 2.672, |
| "grad_norm": 20.75, |
| "learning_rate": 9.85925925925926e-06, |
| "loss": 12.6225, |
| "mean_token_accuracy": 0.7959829289466143, |
| "num_tokens": 47524916.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.7666555680334568, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 17.75, |
| "learning_rate": 9.785185185185187e-06, |
| "loss": 12.3873, |
| "mean_token_accuracy": 0.8007959760725498, |
| "num_tokens": 47807131.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.7808261090889573, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 20.25, |
| "learning_rate": 9.711111111111111e-06, |
| "loss": 12.5291, |
| "mean_token_accuracy": 0.7985376209020615, |
| "num_tokens": 48099654.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.767375604622066, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 18.125, |
| "learning_rate": 9.637037037037038e-06, |
| "loss": 12.3828, |
| "mean_token_accuracy": 0.8003884300589561, |
| "num_tokens": 48375019.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_biology_entropy": 3.164759512901306, |
| "eval_biology_loss": 3.6015591621398926, |
| "eval_biology_mean_token_accuracy": 0.4435784800052643, |
| "eval_biology_num_tokens": 48375019.0, |
| "eval_biology_runtime": 50.0111, |
| "eval_biology_samples_per_second": 9.998, |
| "eval_biology_steps_per_second": 2.499, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_chemistry_entropy": 1.691275239467621, |
| "eval_chemistry_loss": 1.851283311843872, |
| "eval_chemistry_mean_token_accuracy": 0.6341288440227508, |
| "eval_chemistry_num_tokens": 48375019.0, |
| "eval_chemistry_runtime": 61.9361, |
| "eval_chemistry_samples_per_second": 8.073, |
| "eval_chemistry_steps_per_second": 2.018, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_math_entropy": 0.789162187576294, |
| "eval_math_loss": 0.8150354027748108, |
| "eval_math_mean_token_accuracy": 0.7929226984977722, |
| "eval_math_num_tokens": 48375019.0, |
| "eval_math_runtime": 63.5001, |
| "eval_math_samples_per_second": 7.874, |
| "eval_math_steps_per_second": 1.969, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_physics_entropy": 1.351978353023529, |
| "eval_physics_loss": 1.4529732465744019, |
| "eval_physics_mean_token_accuracy": 0.6947901611328124, |
| "eval_physics_num_tokens": 48375019.0, |
| "eval_physics_runtime": 72.3164, |
| "eval_physics_samples_per_second": 6.914, |
| "eval_physics_steps_per_second": 1.729, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.7790494997054338, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 18.5, |
| "learning_rate": 9.562962962962965e-06, |
| "loss": 12.4792, |
| "mean_token_accuracy": 0.7985717471688986, |
| "num_tokens": 48659284.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.764357958547771, |
| "epoch": 2.752, |
| "grad_norm": 16.125, |
| "learning_rate": 9.48888888888889e-06, |
| "loss": 12.3578, |
| "mean_token_accuracy": 0.8015860449522734, |
| "num_tokens": 48943804.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.7782343808561564, |
| "epoch": 2.768, |
| "grad_norm": 17.0, |
| "learning_rate": 9.414814814814816e-06, |
| "loss": 12.4613, |
| "mean_token_accuracy": 0.8002232518047094, |
| "num_tokens": 49230939.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.7899931840598583, |
| "epoch": 2.784, |
| "grad_norm": 18.75, |
| "learning_rate": 9.34074074074074e-06, |
| "loss": 12.7579, |
| "mean_token_accuracy": 0.7950636275112629, |
| "num_tokens": 49504425.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.7675807602703572, |
| "epoch": 2.8, |
| "grad_norm": 14.4375, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 12.438, |
| "mean_token_accuracy": 0.8018883977085351, |
| "num_tokens": 49782661.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.778985939361155, |
| "epoch": 2.816, |
| "grad_norm": 21.0, |
| "learning_rate": 9.192592592592594e-06, |
| "loss": 12.5111, |
| "mean_token_accuracy": 0.7987669218331576, |
| "num_tokens": 50073632.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.7664646361023187, |
| "epoch": 2.832, |
| "grad_norm": 18.5, |
| "learning_rate": 9.118518518518518e-06, |
| "loss": 12.291, |
| "mean_token_accuracy": 0.8016809683293105, |
| "num_tokens": 50356964.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.7853577392175793, |
| "epoch": 2.848, |
| "grad_norm": 15.5625, |
| "learning_rate": 9.044444444444445e-06, |
| "loss": 12.7261, |
| "mean_token_accuracy": 0.7990289311856031, |
| "num_tokens": 50644535.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.7845117872580886, |
| "epoch": 2.864, |
| "grad_norm": 19.75, |
| "learning_rate": 8.970370370370372e-06, |
| "loss": 12.5789, |
| "mean_token_accuracy": 0.7994276065379381, |
| "num_tokens": 50925653.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.7822505347430706, |
| "epoch": 2.88, |
| "grad_norm": 17.875, |
| "learning_rate": 8.896296296296298e-06, |
| "loss": 12.4373, |
| "mean_token_accuracy": 0.8000222198665142, |
| "num_tokens": 51204374.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_biology_entropy": 3.1205562562942504, |
| "eval_biology_loss": 3.597853422164917, |
| "eval_biology_mean_token_accuracy": 0.4442651410102844, |
| "eval_biology_num_tokens": 51204374.0, |
| "eval_biology_runtime": 49.8879, |
| "eval_biology_samples_per_second": 10.022, |
| "eval_biology_steps_per_second": 2.506, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_chemistry_entropy": 1.6736805543899536, |
| "eval_chemistry_loss": 1.8534038066864014, |
| "eval_chemistry_mean_token_accuracy": 0.6339424045085907, |
| "eval_chemistry_num_tokens": 51204374.0, |
| "eval_chemistry_runtime": 62.0043, |
| "eval_chemistry_samples_per_second": 8.064, |
| "eval_chemistry_steps_per_second": 2.016, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_math_entropy": 0.7817594754695892, |
| "eval_math_loss": 0.8139919638633728, |
| "eval_math_mean_token_accuracy": 0.7932133536338806, |
| "eval_math_num_tokens": 51204374.0, |
| "eval_math_runtime": 63.4578, |
| "eval_math_samples_per_second": 7.879, |
| "eval_math_steps_per_second": 1.97, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_physics_entropy": 1.3417515392303467, |
| "eval_physics_loss": 1.4538578987121582, |
| "eval_physics_mean_token_accuracy": 0.6947827172279358, |
| "eval_physics_num_tokens": 51204374.0, |
| "eval_physics_runtime": 72.3122, |
| "eval_physics_samples_per_second": 6.914, |
| "eval_physics_steps_per_second": 1.729, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.7547357704490423, |
| "epoch": 2.896, |
| "grad_norm": 19.375, |
| "learning_rate": 8.822222222222223e-06, |
| "loss": 12.2456, |
| "mean_token_accuracy": 0.802779458835721, |
| "num_tokens": 51483944.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.7747022457420826, |
| "epoch": 2.912, |
| "grad_norm": 16.25, |
| "learning_rate": 8.74814814814815e-06, |
| "loss": 12.4963, |
| "mean_token_accuracy": 0.8008044168353081, |
| "num_tokens": 51765755.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.7698510348796844, |
| "epoch": 2.928, |
| "grad_norm": 18.625, |
| "learning_rate": 8.674074074074074e-06, |
| "loss": 12.3867, |
| "mean_token_accuracy": 0.8028676487505436, |
| "num_tokens": 52056379.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.7608322866261006, |
| "epoch": 2.944, |
| "grad_norm": 15.875, |
| "learning_rate": 8.6e-06, |
| "loss": 12.1279, |
| "mean_token_accuracy": 0.8032333418726921, |
| "num_tokens": 52346232.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.7730306848883629, |
| "epoch": 2.96, |
| "grad_norm": 16.25, |
| "learning_rate": 8.525925925925927e-06, |
| "loss": 12.4642, |
| "mean_token_accuracy": 0.799972441419959, |
| "num_tokens": 52633789.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.7579895993694663, |
| "epoch": 2.976, |
| "grad_norm": 15.8125, |
| "learning_rate": 8.451851851851852e-06, |
| "loss": 12.2556, |
| "mean_token_accuracy": 0.8013919178396464, |
| "num_tokens": 52911755.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.7674846287816763, |
| "epoch": 2.992, |
| "grad_norm": 17.5, |
| "learning_rate": 8.377777777777779e-06, |
| "loss": 12.298, |
| "mean_token_accuracy": 0.8013678915798664, |
| "num_tokens": 53198176.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.7773185564205051, |
| "epoch": 3.008, |
| "grad_norm": 16.875, |
| "learning_rate": 8.303703703703705e-06, |
| "loss": 12.508, |
| "mean_token_accuracy": 0.798418315127492, |
| "num_tokens": 53481893.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.7800862172618508, |
| "epoch": 3.024, |
| "grad_norm": 16.75, |
| "learning_rate": 8.229629629629632e-06, |
| "loss": 12.4563, |
| "mean_token_accuracy": 0.7996578209102154, |
| "num_tokens": 53771717.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.7750446792691946, |
| "epoch": 3.04, |
| "grad_norm": 17.375, |
| "learning_rate": 8.155555555555556e-06, |
| "loss": 12.4342, |
| "mean_token_accuracy": 0.8009418629109859, |
| "num_tokens": 54058045.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_biology_entropy": 3.107721269607544, |
| "eval_biology_loss": 3.5990185737609863, |
| "eval_biology_mean_token_accuracy": 0.4441044263839722, |
| "eval_biology_num_tokens": 54058045.0, |
| "eval_biology_runtime": 49.9074, |
| "eval_biology_samples_per_second": 10.019, |
| "eval_biology_steps_per_second": 2.505, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_chemistry_entropy": 1.6671088938713075, |
| "eval_chemistry_loss": 1.8532856702804565, |
| "eval_chemistry_mean_token_accuracy": 0.6336837713718414, |
| "eval_chemistry_num_tokens": 54058045.0, |
| "eval_chemistry_runtime": 61.8879, |
| "eval_chemistry_samples_per_second": 8.079, |
| "eval_chemistry_steps_per_second": 2.02, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_math_entropy": 0.7754294245243073, |
| "eval_math_loss": 0.813345730304718, |
| "eval_math_mean_token_accuracy": 0.7935237107276917, |
| "eval_math_num_tokens": 54058045.0, |
| "eval_math_runtime": 63.443, |
| "eval_math_samples_per_second": 7.881, |
| "eval_math_steps_per_second": 1.97, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_physics_entropy": 1.334723316669464, |
| "eval_physics_loss": 1.4524449110031128, |
| "eval_physics_mean_token_accuracy": 0.6951661128997803, |
| "eval_physics_num_tokens": 54058045.0, |
| "eval_physics_runtime": 72.2995, |
| "eval_physics_samples_per_second": 6.916, |
| "eval_physics_steps_per_second": 1.729, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.7721475694328547, |
| "epoch": 3.056, |
| "grad_norm": 15.5, |
| "learning_rate": 8.081481481481483e-06, |
| "loss": 12.421, |
| "mean_token_accuracy": 0.8002399168908596, |
| "num_tokens": 54334332.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.753723288141191, |
| "epoch": 3.072, |
| "grad_norm": 16.25, |
| "learning_rate": 8.007407407407408e-06, |
| "loss": 12.135, |
| "mean_token_accuracy": 0.8038679670542479, |
| "num_tokens": 54624543.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.7442250585183501, |
| "epoch": 3.088, |
| "grad_norm": 17.25, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 11.9773, |
| "mean_token_accuracy": 0.8071229167282581, |
| "num_tokens": 54907550.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.7584889115765691, |
| "epoch": 3.104, |
| "grad_norm": 14.625, |
| "learning_rate": 7.859259259259259e-06, |
| "loss": 12.1175, |
| "mean_token_accuracy": 0.8044587153941393, |
| "num_tokens": 55190959.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.7494216404855252, |
| "epoch": 3.12, |
| "grad_norm": 18.625, |
| "learning_rate": 7.785185185185185e-06, |
| "loss": 11.9723, |
| "mean_token_accuracy": 0.8067513905465603, |
| "num_tokens": 55481635.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.7434225969016552, |
| "epoch": 3.136, |
| "grad_norm": 16.375, |
| "learning_rate": 7.711111111111112e-06, |
| "loss": 12.0611, |
| "mean_token_accuracy": 0.8049594823271036, |
| "num_tokens": 55769010.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.762983326241374, |
| "epoch": 3.152, |
| "grad_norm": 16.625, |
| "learning_rate": 7.637037037037037e-06, |
| "loss": 12.2437, |
| "mean_token_accuracy": 0.8035841189324856, |
| "num_tokens": 56053160.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.77687908988446, |
| "epoch": 3.168, |
| "grad_norm": 16.0, |
| "learning_rate": 7.562962962962963e-06, |
| "loss": 12.4755, |
| "mean_token_accuracy": 0.8001125130802393, |
| "num_tokens": 56337066.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.7547883773222566, |
| "epoch": 3.184, |
| "grad_norm": 18.375, |
| "learning_rate": 7.48888888888889e-06, |
| "loss": 12.0706, |
| "mean_token_accuracy": 0.8026409812271595, |
| "num_tokens": 56618899.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.7374519377946853, |
| "epoch": 3.2, |
| "grad_norm": 14.1875, |
| "learning_rate": 7.4148148148148155e-06, |
| "loss": 11.8554, |
| "mean_token_accuracy": 0.8069235980510712, |
| "num_tokens": 56910071.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_biology_entropy": 3.0970921745300295, |
| "eval_biology_loss": 3.6141059398651123, |
| "eval_biology_mean_token_accuracy": 0.44385591268539426, |
| "eval_biology_num_tokens": 56910071.0, |
| "eval_biology_runtime": 49.9375, |
| "eval_biology_samples_per_second": 10.013, |
| "eval_biology_steps_per_second": 2.503, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_chemistry_entropy": 1.660459312915802, |
| "eval_chemistry_loss": 1.8578524589538574, |
| "eval_chemistry_mean_token_accuracy": 0.6334774570465088, |
| "eval_chemistry_num_tokens": 56910071.0, |
| "eval_chemistry_runtime": 61.6056, |
| "eval_chemistry_samples_per_second": 8.116, |
| "eval_chemistry_steps_per_second": 2.029, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_math_entropy": 0.7726022355556488, |
| "eval_math_loss": 0.8128476738929749, |
| "eval_math_mean_token_accuracy": 0.7935845293998718, |
| "eval_math_num_tokens": 56910071.0, |
| "eval_math_runtime": 63.3659, |
| "eval_math_samples_per_second": 7.891, |
| "eval_math_steps_per_second": 1.973, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_physics_entropy": 1.3269246454238892, |
| "eval_physics_loss": 1.456142544746399, |
| "eval_physics_mean_token_accuracy": 0.6948826293945313, |
| "eval_physics_num_tokens": 56910071.0, |
| "eval_physics_runtime": 72.2501, |
| "eval_physics_samples_per_second": 6.92, |
| "eval_physics_steps_per_second": 1.73, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 3000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "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": 5.791878293573609e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
