| { |
| "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": 1.0476444251835346, |
| "epoch": 0.016, |
| "grad_norm": 158.0, |
| "learning_rate": 6.000000000000001e-07, |
| "loss": 34.7404, |
| "mean_token_accuracy": 0.6322720043361187, |
| "num_tokens": 280941.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.103129617869854, |
| "epoch": 0.032, |
| "grad_norm": 140.0, |
| "learning_rate": 1.2666666666666669e-06, |
| "loss": 35.2833, |
| "mean_token_accuracy": 0.627065640874207, |
| "num_tokens": 558056.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.1186710916459561, |
| "epoch": 0.048, |
| "grad_norm": 125.5, |
| "learning_rate": 1.9333333333333336e-06, |
| "loss": 34.279, |
| "mean_token_accuracy": 0.6317529268562794, |
| "num_tokens": 836753.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.2232312321662904, |
| "epoch": 0.064, |
| "grad_norm": 99.0, |
| "learning_rate": 2.6e-06, |
| "loss": 33.2898, |
| "mean_token_accuracy": 0.6292566264048218, |
| "num_tokens": 1126446.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.3642345190048217, |
| "epoch": 0.08, |
| "grad_norm": 78.0, |
| "learning_rate": 3.266666666666667e-06, |
| "loss": 31.1081, |
| "mean_token_accuracy": 0.643590996041894, |
| "num_tokens": 1413596.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.4666775342077016, |
| "epoch": 0.096, |
| "grad_norm": 50.0, |
| "learning_rate": 3.9333333333333335e-06, |
| "loss": 28.0855, |
| "mean_token_accuracy": 0.6677330181002616, |
| "num_tokens": 1701193.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.479279049113393, |
| "epoch": 0.112, |
| "grad_norm": 44.5, |
| "learning_rate": 4.600000000000001e-06, |
| "loss": 25.9511, |
| "mean_token_accuracy": 0.6763931937515736, |
| "num_tokens": 1979232.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.4356445044279098, |
| "epoch": 0.128, |
| "grad_norm": 31.5, |
| "learning_rate": 5.2666666666666665e-06, |
| "loss": 24.2383, |
| "mean_token_accuracy": 0.6900208964943886, |
| "num_tokens": 2274177.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.3773791242390871, |
| "epoch": 0.144, |
| "grad_norm": 28.25, |
| "learning_rate": 5.933333333333335e-06, |
| "loss": 22.3128, |
| "mean_token_accuracy": 0.7042559664696455, |
| "num_tokens": 2548445.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.3055105797946454, |
| "epoch": 0.16, |
| "grad_norm": 29.25, |
| "learning_rate": 6.600000000000001e-06, |
| "loss": 20.6858, |
| "mean_token_accuracy": 0.718151243403554, |
| "num_tokens": 2824418.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 3.5783245754241944, |
| "eval_biology_loss": 4.867801189422607, |
| "eval_biology_mean_token_accuracy": 0.3678655695915222, |
| "eval_biology_num_tokens": 2824418.0, |
| "eval_biology_runtime": 50.0015, |
| "eval_biology_samples_per_second": 10.0, |
| "eval_biology_steps_per_second": 2.5, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 2.541287685394287, |
| "eval_chemistry_loss": 2.7937169075012207, |
| "eval_chemistry_mean_token_accuracy": 0.539857518196106, |
| "eval_chemistry_num_tokens": 2824418.0, |
| "eval_chemistry_runtime": 61.8974, |
| "eval_chemistry_samples_per_second": 8.078, |
| "eval_chemistry_steps_per_second": 2.019, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 1.325653440952301, |
| "eval_math_loss": 1.2963088750839233, |
| "eval_math_mean_token_accuracy": 0.7152829113006591, |
| "eval_math_num_tokens": 2824418.0, |
| "eval_math_runtime": 63.4831, |
| "eval_math_samples_per_second": 7.876, |
| "eval_math_steps_per_second": 1.969, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 2.0472509851455687, |
| "eval_physics_loss": 2.108372688293457, |
| "eval_physics_mean_token_accuracy": 0.6150235052108765, |
| "eval_physics_num_tokens": 2824418.0, |
| "eval_physics_runtime": 72.2792, |
| "eval_physics_samples_per_second": 6.918, |
| "eval_physics_steps_per_second": 1.729, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.2799372777342797, |
| "epoch": 0.176, |
| "grad_norm": 23.25, |
| "learning_rate": 7.266666666666668e-06, |
| "loss": 20.1214, |
| "mean_token_accuracy": 0.7223256763070822, |
| "num_tokens": 3110313.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.271880804002285, |
| "epoch": 0.192, |
| "grad_norm": 22.625, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 19.8735, |
| "mean_token_accuracy": 0.7208763293921947, |
| "num_tokens": 3394170.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.2141485448926688, |
| "epoch": 0.208, |
| "grad_norm": 20.25, |
| "learning_rate": 8.6e-06, |
| "loss": 19.3285, |
| "mean_token_accuracy": 0.7284529402852058, |
| "num_tokens": 3673600.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.1842644162476064, |
| "epoch": 0.224, |
| "grad_norm": 22.25, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 18.8222, |
| "mean_token_accuracy": 0.7318667653948069, |
| "num_tokens": 3953732.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.1367911960929633, |
| "epoch": 0.24, |
| "grad_norm": 21.875, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 18.0331, |
| "mean_token_accuracy": 0.7411410234868526, |
| "num_tokens": 4243655.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.099366496130824, |
| "epoch": 0.256, |
| "grad_norm": 20.875, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 17.5038, |
| "mean_token_accuracy": 0.7470747366547584, |
| "num_tokens": 4531471.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.1072874456644057, |
| "epoch": 0.272, |
| "grad_norm": 18.125, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 17.6358, |
| "mean_token_accuracy": 0.7453399378806352, |
| "num_tokens": 4810284.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.0879811342805623, |
| "epoch": 0.288, |
| "grad_norm": 16.375, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 17.2142, |
| "mean_token_accuracy": 0.749498899653554, |
| "num_tokens": 5095104.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.0824080746620894, |
| "epoch": 0.304, |
| "grad_norm": 16.625, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 17.223, |
| "mean_token_accuracy": 0.7493264392018318, |
| "num_tokens": 5383732.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.048988512158394, |
| "epoch": 0.32, |
| "grad_norm": 19.125, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 16.7231, |
| "mean_token_accuracy": 0.7555124312639236, |
| "num_tokens": 5676334.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 3.5249868144989014, |
| "eval_biology_loss": 4.3318328857421875, |
| "eval_biology_mean_token_accuracy": 0.39426790857315064, |
| "eval_biology_num_tokens": 5676334.0, |
| "eval_biology_runtime": 49.8836, |
| "eval_biology_samples_per_second": 10.023, |
| "eval_biology_steps_per_second": 2.506, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 2.3051584205627442, |
| "eval_chemistry_loss": 2.4184799194335938, |
| "eval_chemistry_mean_token_accuracy": 0.5702217166423798, |
| "eval_chemistry_num_tokens": 5676334.0, |
| "eval_chemistry_runtime": 61.8203, |
| "eval_chemistry_samples_per_second": 8.088, |
| "eval_chemistry_steps_per_second": 2.022, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 1.0760114483833314, |
| "eval_math_loss": 1.0703883171081543, |
| "eval_math_mean_token_accuracy": 0.7492856097221374, |
| "eval_math_num_tokens": 5676334.0, |
| "eval_math_runtime": 63.3918, |
| "eval_math_samples_per_second": 7.887, |
| "eval_math_steps_per_second": 1.972, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 1.7677014241218567, |
| "eval_physics_loss": 1.8187588453292847, |
| "eval_physics_mean_token_accuracy": 0.6450894718170166, |
| "eval_physics_num_tokens": 5676334.0, |
| "eval_physics_runtime": 72.1893, |
| "eval_physics_samples_per_second": 6.926, |
| "eval_physics_steps_per_second": 1.732, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.0659823317080737, |
| "epoch": 0.336, |
| "grad_norm": 18.125, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 17.0855, |
| "mean_token_accuracy": 0.7487949147820473, |
| "num_tokens": 5958480.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.0319596778601408, |
| "epoch": 0.352, |
| "grad_norm": 16.625, |
| "learning_rate": 1.46e-05, |
| "loss": 16.3603, |
| "mean_token_accuracy": 0.7570414818823338, |
| "num_tokens": 6242161.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 1.0307059772312641, |
| "epoch": 0.368, |
| "grad_norm": 19.125, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 16.4803, |
| "mean_token_accuracy": 0.7558304250240326, |
| "num_tokens": 6523679.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.008485846966505, |
| "epoch": 0.384, |
| "grad_norm": 17.75, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 16.1119, |
| "mean_token_accuracy": 0.7602646600455045, |
| "num_tokens": 6810978.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.0437373481690884, |
| "epoch": 0.4, |
| "grad_norm": 17.125, |
| "learning_rate": 1.66e-05, |
| "loss": 16.936, |
| "mean_token_accuracy": 0.7534951154142618, |
| "num_tokens": 7096903.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 1.0108945474028588, |
| "epoch": 0.416, |
| "grad_norm": 16.75, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 16.165, |
| "mean_token_accuracy": 0.7585977960377932, |
| "num_tokens": 7377181.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.9808746818453074, |
| "epoch": 0.432, |
| "grad_norm": 19.125, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 15.708, |
| "mean_token_accuracy": 0.7660888392478228, |
| "num_tokens": 7650523.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.9656771093606948, |
| "epoch": 0.448, |
| "grad_norm": 19.375, |
| "learning_rate": 1.86e-05, |
| "loss": 15.4934, |
| "mean_token_accuracy": 0.7665207251906395, |
| "num_tokens": 7936788.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.970192850753665, |
| "epoch": 0.464, |
| "grad_norm": 21.125, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 15.697, |
| "mean_token_accuracy": 0.7657037891447545, |
| "num_tokens": 8223147.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.9936408184468746, |
| "epoch": 0.48, |
| "grad_norm": 18.375, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 15.8279, |
| "mean_token_accuracy": 0.7644913751631975, |
| "num_tokens": 8506030.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 3.547258296966553, |
| "eval_biology_loss": 4.036286354064941, |
| "eval_biology_mean_token_accuracy": 0.4092668824195862, |
| "eval_biology_num_tokens": 8506030.0, |
| "eval_biology_runtime": 49.8646, |
| "eval_biology_samples_per_second": 10.027, |
| "eval_biology_steps_per_second": 2.507, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 2.243741189956665, |
| "eval_chemistry_loss": 2.2735037803649902, |
| "eval_chemistry_mean_token_accuracy": 0.585998829126358, |
| "eval_chemistry_num_tokens": 8506030.0, |
| "eval_chemistry_runtime": 61.8087, |
| "eval_chemistry_samples_per_second": 8.089, |
| "eval_chemistry_steps_per_second": 2.022, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 0.969578085899353, |
| "eval_math_loss": 0.9849457740783691, |
| "eval_math_mean_token_accuracy": 0.7633716745376586, |
| "eval_math_num_tokens": 8506030.0, |
| "eval_math_runtime": 63.4567, |
| "eval_math_samples_per_second": 7.879, |
| "eval_math_steps_per_second": 1.97, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 1.6970900883674622, |
| "eval_physics_loss": 1.7149206399917603, |
| "eval_physics_mean_token_accuracy": 0.6588896939754486, |
| "eval_physics_num_tokens": 8506030.0, |
| "eval_physics_runtime": 72.1904, |
| "eval_physics_samples_per_second": 6.926, |
| "eval_physics_steps_per_second": 1.732, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.9509127460420131, |
| "epoch": 0.496, |
| "grad_norm": 27.125, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 15.2509, |
| "mean_token_accuracy": 0.7699485655874014, |
| "num_tokens": 8788726.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.9583811838179827, |
| "epoch": 0.512, |
| "grad_norm": 17.75, |
| "learning_rate": 1.985925925925926e-05, |
| "loss": 15.3156, |
| "mean_token_accuracy": 0.7701035752892494, |
| "num_tokens": 9078039.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.9286689113825559, |
| "epoch": 0.528, |
| "grad_norm": 21.375, |
| "learning_rate": 1.9785185185185187e-05, |
| "loss": 14.9975, |
| "mean_token_accuracy": 0.7747602745890617, |
| "num_tokens": 9373860.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.93574563190341, |
| "epoch": 0.544, |
| "grad_norm": 19.0, |
| "learning_rate": 1.971111111111111e-05, |
| "loss": 15.1982, |
| "mean_token_accuracy": 0.7723268665373325, |
| "num_tokens": 9660940.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.9833962079137564, |
| "epoch": 0.56, |
| "grad_norm": 27.375, |
| "learning_rate": 1.963703703703704e-05, |
| "loss": 15.6566, |
| "mean_token_accuracy": 0.7635074742138386, |
| "num_tokens": 9932302.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.9380292896181345, |
| "epoch": 0.576, |
| "grad_norm": 19.5, |
| "learning_rate": 1.9562962962962964e-05, |
| "loss": 15.0896, |
| "mean_token_accuracy": 0.771076611801982, |
| "num_tokens": 10215462.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.9226088404655457, |
| "epoch": 0.592, |
| "grad_norm": 17.125, |
| "learning_rate": 1.948888888888889e-05, |
| "loss": 14.9585, |
| "mean_token_accuracy": 0.7726155135780572, |
| "num_tokens": 10504396.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.9383664086461068, |
| "epoch": 0.608, |
| "grad_norm": 16.875, |
| "learning_rate": 1.9414814814814817e-05, |
| "loss": 15.0747, |
| "mean_token_accuracy": 0.772657984867692, |
| "num_tokens": 10793126.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.9145499892532826, |
| "epoch": 0.624, |
| "grad_norm": 18.875, |
| "learning_rate": 1.9340740740740743e-05, |
| "loss": 14.7611, |
| "mean_token_accuracy": 0.7738783087581396, |
| "num_tokens": 11081768.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.9076093930751086, |
| "epoch": 0.64, |
| "grad_norm": 17.5, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 14.491, |
| "mean_token_accuracy": 0.7761218074709177, |
| "num_tokens": 11370320.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 3.4644125595092774, |
| "eval_biology_loss": 3.9791781902313232, |
| "eval_biology_mean_token_accuracy": 0.4146076333522797, |
| "eval_biology_num_tokens": 11370320.0, |
| "eval_biology_runtime": 49.8972, |
| "eval_biology_samples_per_second": 10.021, |
| "eval_biology_steps_per_second": 2.505, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 2.1470180315971374, |
| "eval_chemistry_loss": 2.1781718730926514, |
| "eval_chemistry_mean_token_accuracy": 0.5962020962238311, |
| "eval_chemistry_num_tokens": 11370320.0, |
| "eval_chemistry_runtime": 61.8451, |
| "eval_chemistry_samples_per_second": 8.085, |
| "eval_chemistry_steps_per_second": 2.021, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 0.9346609992980957, |
| "eval_math_loss": 0.9384942054748535, |
| "eval_math_mean_token_accuracy": 0.771290735244751, |
| "eval_math_num_tokens": 11370320.0, |
| "eval_math_runtime": 63.3885, |
| "eval_math_samples_per_second": 7.888, |
| "eval_math_steps_per_second": 1.972, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 1.6158739218711853, |
| "eval_physics_loss": 1.6468137502670288, |
| "eval_physics_mean_token_accuracy": 0.6678030126094818, |
| "eval_physics_num_tokens": 11370320.0, |
| "eval_physics_runtime": 72.1919, |
| "eval_physics_samples_per_second": 6.926, |
| "eval_physics_steps_per_second": 1.731, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.8844340030103922, |
| "epoch": 0.656, |
| "grad_norm": 15.625, |
| "learning_rate": 1.9192592592592593e-05, |
| "loss": 14.2919, |
| "mean_token_accuracy": 0.7790201846510172, |
| "num_tokens": 11657835.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.9230461683124304, |
| "epoch": 0.672, |
| "grad_norm": 17.25, |
| "learning_rate": 1.911851851851852e-05, |
| "loss": 14.8067, |
| "mean_token_accuracy": 0.7730407487601042, |
| "num_tokens": 11949262.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.8864722024649382, |
| "epoch": 0.688, |
| "grad_norm": 14.5625, |
| "learning_rate": 1.9044444444444446e-05, |
| "loss": 14.2543, |
| "mean_token_accuracy": 0.7801163170486689, |
| "num_tokens": 12227640.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.8921929065138101, |
| "epoch": 0.704, |
| "grad_norm": 17.5, |
| "learning_rate": 1.8970370370370372e-05, |
| "loss": 14.3854, |
| "mean_token_accuracy": 0.7790493600070476, |
| "num_tokens": 12516641.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.9114352393895387, |
| "epoch": 0.72, |
| "grad_norm": 18.75, |
| "learning_rate": 1.8896296296296295e-05, |
| "loss": 14.7549, |
| "mean_token_accuracy": 0.7744539573788642, |
| "num_tokens": 12793343.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.9004437670111656, |
| "epoch": 0.736, |
| "grad_norm": 16.625, |
| "learning_rate": 1.8822222222222225e-05, |
| "loss": 14.4296, |
| "mean_token_accuracy": 0.7780555929988623, |
| "num_tokens": 13077981.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.8800053089857102, |
| "epoch": 0.752, |
| "grad_norm": 19.5, |
| "learning_rate": 1.874814814814815e-05, |
| "loss": 14.1044, |
| "mean_token_accuracy": 0.7824321106076241, |
| "num_tokens": 13358957.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.8506957441568375, |
| "epoch": 0.768, |
| "grad_norm": 15.25, |
| "learning_rate": 1.8674074074074075e-05, |
| "loss": 13.7613, |
| "mean_token_accuracy": 0.7884838156402111, |
| "num_tokens": 13653412.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.9173242174088955, |
| "epoch": 0.784, |
| "grad_norm": 18.125, |
| "learning_rate": 1.86e-05, |
| "loss": 14.7366, |
| "mean_token_accuracy": 0.776311819255352, |
| "num_tokens": 13940856.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.9153516864404082, |
| "epoch": 0.8, |
| "grad_norm": 18.5, |
| "learning_rate": 1.8525925925925928e-05, |
| "loss": 14.84, |
| "mean_token_accuracy": 0.7749685924500227, |
| "num_tokens": 14230754.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 3.587346399307251, |
| "eval_biology_loss": 3.8396143913269043, |
| "eval_biology_mean_token_accuracy": 0.42098368620872495, |
| "eval_biology_num_tokens": 14230754.0, |
| "eval_biology_runtime": 49.8808, |
| "eval_biology_samples_per_second": 10.024, |
| "eval_biology_steps_per_second": 2.506, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 2.168390507221222, |
| "eval_chemistry_loss": 2.1044840812683105, |
| "eval_chemistry_mean_token_accuracy": 0.6037136344909668, |
| "eval_chemistry_num_tokens": 14230754.0, |
| "eval_chemistry_runtime": 61.4676, |
| "eval_chemistry_samples_per_second": 8.134, |
| "eval_chemistry_steps_per_second": 2.034, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 0.9173241977691651, |
| "eval_math_loss": 0.911715030670166, |
| "eval_math_mean_token_accuracy": 0.7763807888031006, |
| "eval_math_num_tokens": 14230754.0, |
| "eval_math_runtime": 63.353, |
| "eval_math_samples_per_second": 7.892, |
| "eval_math_steps_per_second": 1.973, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 1.6311655960083007, |
| "eval_physics_loss": 1.610815167427063, |
| "eval_physics_mean_token_accuracy": 0.6733945074081421, |
| "eval_physics_num_tokens": 14230754.0, |
| "eval_physics_runtime": 72.3928, |
| "eval_physics_samples_per_second": 6.907, |
| "eval_physics_steps_per_second": 1.727, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.8935465440154076, |
| "epoch": 0.816, |
| "grad_norm": 16.875, |
| "learning_rate": 1.8451851851851855e-05, |
| "loss": 14.1977, |
| "mean_token_accuracy": 0.7802476394921541, |
| "num_tokens": 14519893.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.8838269270956516, |
| "epoch": 0.832, |
| "grad_norm": 17.375, |
| "learning_rate": 1.8377777777777778e-05, |
| "loss": 14.2483, |
| "mean_token_accuracy": 0.7819081086665391, |
| "num_tokens": 14805088.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.9093101542443037, |
| "epoch": 0.848, |
| "grad_norm": 18.125, |
| "learning_rate": 1.8303703703703704e-05, |
| "loss": 14.6777, |
| "mean_token_accuracy": 0.7753469321876765, |
| "num_tokens": 15086306.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.900922878831625, |
| "epoch": 0.864, |
| "grad_norm": 15.0, |
| "learning_rate": 1.822962962962963e-05, |
| "loss": 14.4243, |
| "mean_token_accuracy": 0.778496702387929, |
| "num_tokens": 15370985.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.9037761628627777, |
| "epoch": 0.88, |
| "grad_norm": 20.5, |
| "learning_rate": 1.8155555555555557e-05, |
| "loss": 14.5507, |
| "mean_token_accuracy": 0.7772340852767229, |
| "num_tokens": 15653836.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.8908271692693234, |
| "epoch": 0.896, |
| "grad_norm": 21.375, |
| "learning_rate": 1.8081481481481484e-05, |
| "loss": 14.3982, |
| "mean_token_accuracy": 0.7781253006309271, |
| "num_tokens": 15932179.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.8755331316962838, |
| "epoch": 0.912, |
| "grad_norm": 17.5, |
| "learning_rate": 1.800740740740741e-05, |
| "loss": 13.9566, |
| "mean_token_accuracy": 0.7827421851456166, |
| "num_tokens": 16219842.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.8637437466531992, |
| "epoch": 0.928, |
| "grad_norm": 19.875, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 13.9502, |
| "mean_token_accuracy": 0.7828048501163721, |
| "num_tokens": 16499842.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.8896576169878244, |
| "epoch": 0.944, |
| "grad_norm": 35.5, |
| "learning_rate": 1.785925925925926e-05, |
| "loss": 14.3781, |
| "mean_token_accuracy": 0.7790289539843798, |
| "num_tokens": 16781882.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.9052465215325356, |
| "epoch": 0.96, |
| "grad_norm": 15.4375, |
| "learning_rate": 1.7785185185185186e-05, |
| "loss": 14.4585, |
| "mean_token_accuracy": 0.7787603437900543, |
| "num_tokens": 17067407.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 3.393627851486206, |
| "eval_biology_loss": 3.804689407348633, |
| "eval_biology_mean_token_accuracy": 0.42461460971832277, |
| "eval_biology_num_tokens": 17067407.0, |
| "eval_biology_runtime": 49.7301, |
| "eval_biology_samples_per_second": 10.054, |
| "eval_biology_steps_per_second": 2.514, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_chemistry_entropy": 2.023163953304291, |
| "eval_chemistry_loss": 2.059583902359009, |
| "eval_chemistry_mean_token_accuracy": 0.608015153169632, |
| "eval_chemistry_num_tokens": 17067407.0, |
| "eval_chemistry_runtime": 61.6301, |
| "eval_chemistry_samples_per_second": 8.113, |
| "eval_chemistry_steps_per_second": 2.028, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_math_entropy": 0.886635826587677, |
| "eval_math_loss": 0.8929259777069092, |
| "eval_math_mean_token_accuracy": 0.7791896166801453, |
| "eval_math_num_tokens": 17067407.0, |
| "eval_math_runtime": 63.1881, |
| "eval_math_samples_per_second": 7.913, |
| "eval_math_steps_per_second": 1.978, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_physics_entropy": 1.5457578830718994, |
| "eval_physics_loss": 1.5915517807006836, |
| "eval_physics_mean_token_accuracy": 0.6765441355705262, |
| "eval_physics_num_tokens": 17067407.0, |
| "eval_physics_runtime": 71.9901, |
| "eval_physics_samples_per_second": 6.945, |
| "eval_physics_steps_per_second": 1.736, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.8741790272295475, |
| "epoch": 0.976, |
| "grad_norm": 20.125, |
| "learning_rate": 1.7711111111111113e-05, |
| "loss": 14.0666, |
| "mean_token_accuracy": 0.7843787413090467, |
| "num_tokens": 17350994.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.8895185798406601, |
| "epoch": 0.992, |
| "grad_norm": 16.875, |
| "learning_rate": 1.763703703703704e-05, |
| "loss": 14.2795, |
| "mean_token_accuracy": 0.7792422734200954, |
| "num_tokens": 17637514.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.8725822484120727, |
| "epoch": 1.008, |
| "grad_norm": 16.25, |
| "learning_rate": 1.7562962962962962e-05, |
| "loss": 13.9234, |
| "mean_token_accuracy": 0.7825319610536099, |
| "num_tokens": 17926570.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.859299965761602, |
| "epoch": 1.024, |
| "grad_norm": 16.375, |
| "learning_rate": 1.7488888888888892e-05, |
| "loss": 13.7487, |
| "mean_token_accuracy": 0.7849128067493438, |
| "num_tokens": 18207652.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.8529570057988167, |
| "epoch": 1.04, |
| "grad_norm": 18.5, |
| "learning_rate": 1.7414814814814815e-05, |
| "loss": 13.7357, |
| "mean_token_accuracy": 0.7864760752767325, |
| "num_tokens": 18484931.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.8539858607575297, |
| "epoch": 1.056, |
| "grad_norm": 18.0, |
| "learning_rate": 1.7340740740740742e-05, |
| "loss": 13.8974, |
| "mean_token_accuracy": 0.7846961792558431, |
| "num_tokens": 18773457.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.8639433614909648, |
| "epoch": 1.072, |
| "grad_norm": 17.625, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 13.7869, |
| "mean_token_accuracy": 0.7835282500833273, |
| "num_tokens": 19055365.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.8525660615414381, |
| "epoch": 1.088, |
| "grad_norm": 15.875, |
| "learning_rate": 1.7192592592592595e-05, |
| "loss": 13.7686, |
| "mean_token_accuracy": 0.786788547411561, |
| "num_tokens": 19345730.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.8591295637190342, |
| "epoch": 1.104, |
| "grad_norm": 14.4375, |
| "learning_rate": 1.711851851851852e-05, |
| "loss": 13.7706, |
| "mean_token_accuracy": 0.7860686339437961, |
| "num_tokens": 19637390.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.8316973434761167, |
| "epoch": 1.12, |
| "grad_norm": 18.0, |
| "learning_rate": 1.7044444444444445e-05, |
| "loss": 13.3392, |
| "mean_token_accuracy": 0.7907124478369951, |
| "num_tokens": 19923914.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_biology_entropy": 3.372835781097412, |
| "eval_biology_loss": 3.8594861030578613, |
| "eval_biology_mean_token_accuracy": 0.42449810791015624, |
| "eval_biology_num_tokens": 19923914.0, |
| "eval_biology_runtime": 49.7902, |
| "eval_biology_samples_per_second": 10.042, |
| "eval_biology_steps_per_second": 2.511, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_chemistry_entropy": 1.985270866394043, |
| "eval_chemistry_loss": 2.0690011978149414, |
| "eval_chemistry_mean_token_accuracy": 0.6098231360912323, |
| "eval_chemistry_num_tokens": 19923914.0, |
| "eval_chemistry_runtime": 61.935, |
| "eval_chemistry_samples_per_second": 8.073, |
| "eval_chemistry_steps_per_second": 2.018, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_math_entropy": 0.8660009272098541, |
| "eval_math_loss": 0.8819393515586853, |
| "eval_math_mean_token_accuracy": 0.7814530291557312, |
| "eval_math_num_tokens": 19923914.0, |
| "eval_math_runtime": 63.377, |
| "eval_math_samples_per_second": 7.889, |
| "eval_math_steps_per_second": 1.972, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_physics_entropy": 1.5101813049316406, |
| "eval_physics_loss": 1.5800307989120483, |
| "eval_physics_mean_token_accuracy": 0.6781299662590027, |
| "eval_physics_num_tokens": 19923914.0, |
| "eval_physics_runtime": 72.2227, |
| "eval_physics_samples_per_second": 6.923, |
| "eval_physics_steps_per_second": 1.731, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.8278432639315725, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 16.875, |
| "learning_rate": 1.697037037037037e-05, |
| "loss": 13.3745, |
| "mean_token_accuracy": 0.7893257014453411, |
| "num_tokens": 20201892.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.8526920214295387, |
| "epoch": 1.152, |
| "grad_norm": 19.0, |
| "learning_rate": 1.6896296296296298e-05, |
| "loss": 13.6557, |
| "mean_token_accuracy": 0.7878582254052162, |
| "num_tokens": 20490282.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.8450694579631091, |
| "epoch": 1.168, |
| "grad_norm": 15.875, |
| "learning_rate": 1.6822222222222224e-05, |
| "loss": 13.6233, |
| "mean_token_accuracy": 0.787573279067874, |
| "num_tokens": 20785786.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.8410906136035919, |
| "epoch": 1.184, |
| "grad_norm": 19.5, |
| "learning_rate": 1.6748148148148147e-05, |
| "loss": 13.5188, |
| "mean_token_accuracy": 0.7897454522550106, |
| "num_tokens": 21074205.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.8331539599224925, |
| "epoch": 1.2, |
| "grad_norm": 18.625, |
| "learning_rate": 1.6674074074074077e-05, |
| "loss": 13.3771, |
| "mean_token_accuracy": 0.7902466226369143, |
| "num_tokens": 21369159.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.8552144020795822, |
| "epoch": 1.216, |
| "grad_norm": 22.25, |
| "learning_rate": 1.66e-05, |
| "loss": 13.7361, |
| "mean_token_accuracy": 0.7870386108756066, |
| "num_tokens": 21649178.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.8264798279851675, |
| "epoch": 1.232, |
| "grad_norm": 18.25, |
| "learning_rate": 1.6525925925925927e-05, |
| "loss": 13.2327, |
| "mean_token_accuracy": 0.7920120354741812, |
| "num_tokens": 21930239.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.8188742805272341, |
| "epoch": 1.248, |
| "grad_norm": 16.625, |
| "learning_rate": 1.6451851851851853e-05, |
| "loss": 13.1839, |
| "mean_token_accuracy": 0.7921639878302813, |
| "num_tokens": 22216387.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.8594657603651286, |
| "epoch": 1.264, |
| "grad_norm": 19.125, |
| "learning_rate": 1.637777777777778e-05, |
| "loss": 13.88, |
| "mean_token_accuracy": 0.7837325409054756, |
| "num_tokens": 22501002.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.8564920598641038, |
| "epoch": 1.28, |
| "grad_norm": 21.375, |
| "learning_rate": 1.6303703703703706e-05, |
| "loss": 13.6499, |
| "mean_token_accuracy": 0.7877908434718848, |
| "num_tokens": 22779682.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_biology_entropy": 3.376770601272583, |
| "eval_biology_loss": 3.801596164703369, |
| "eval_biology_mean_token_accuracy": 0.4262664563655853, |
| "eval_biology_num_tokens": 22779682.0, |
| "eval_biology_runtime": 49.8531, |
| "eval_biology_samples_per_second": 10.029, |
| "eval_biology_steps_per_second": 2.507, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_chemistry_entropy": 1.9474885282516479, |
| "eval_chemistry_loss": 2.0175445079803467, |
| "eval_chemistry_mean_token_accuracy": 0.6140047969818115, |
| "eval_chemistry_num_tokens": 22779682.0, |
| "eval_chemistry_runtime": 61.8009, |
| "eval_chemistry_samples_per_second": 8.09, |
| "eval_chemistry_steps_per_second": 2.023, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_math_entropy": 0.839567274093628, |
| "eval_math_loss": 0.8709415197372437, |
| "eval_math_mean_token_accuracy": 0.783750717163086, |
| "eval_math_num_tokens": 22779682.0, |
| "eval_math_runtime": 63.3865, |
| "eval_math_samples_per_second": 7.888, |
| "eval_math_steps_per_second": 1.972, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_physics_entropy": 1.488646312236786, |
| "eval_physics_loss": 1.5581644773483276, |
| "eval_physics_mean_token_accuracy": 0.6805251049995422, |
| "eval_physics_num_tokens": 22779682.0, |
| "eval_physics_runtime": 72.4407, |
| "eval_physics_samples_per_second": 6.902, |
| "eval_physics_steps_per_second": 1.726, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.8242643151432276, |
| "epoch": 1.296, |
| "grad_norm": 16.75, |
| "learning_rate": 1.622962962962963e-05, |
| "loss": 13.3485, |
| "mean_token_accuracy": 0.7901775784790516, |
| "num_tokens": 23057744.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.8360366895794868, |
| "epoch": 1.312, |
| "grad_norm": 20.25, |
| "learning_rate": 1.6155555555555556e-05, |
| "loss": 13.6015, |
| "mean_token_accuracy": 0.7880489483475686, |
| "num_tokens": 23344644.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.8606906643137335, |
| "epoch": 1.328, |
| "grad_norm": 19.0, |
| "learning_rate": 1.6081481481481482e-05, |
| "loss": 13.6974, |
| "mean_token_accuracy": 0.7855447053909301, |
| "num_tokens": 23622405.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.8278559835627675, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 16.5, |
| "learning_rate": 1.600740740740741e-05, |
| "loss": 13.2704, |
| "mean_token_accuracy": 0.7906895779073239, |
| "num_tokens": 23899771.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.8220528084784746, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 15.5625, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 13.3103, |
| "mean_token_accuracy": 0.7912323564291001, |
| "num_tokens": 24187023.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.8924444895237684, |
| "epoch": 1.376, |
| "grad_norm": 20.875, |
| "learning_rate": 1.5859259259259262e-05, |
| "loss": 14.3535, |
| "mean_token_accuracy": 0.7774244502186776, |
| "num_tokens": 24466132.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.8332192454487085, |
| "epoch": 1.392, |
| "grad_norm": 19.5, |
| "learning_rate": 1.5785185185185185e-05, |
| "loss": 13.3708, |
| "mean_token_accuracy": 0.7895895481109619, |
| "num_tokens": 24748043.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.816296424344182, |
| "epoch": 1.408, |
| "grad_norm": 19.75, |
| "learning_rate": 1.571111111111111e-05, |
| "loss": 13.2019, |
| "mean_token_accuracy": 0.7928898848593235, |
| "num_tokens": 25036674.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.81595440171659, |
| "epoch": 1.424, |
| "grad_norm": 14.375, |
| "learning_rate": 1.5637037037037038e-05, |
| "loss": 13.1288, |
| "mean_token_accuracy": 0.793658908829093, |
| "num_tokens": 25324579.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.8419610138982534, |
| "epoch": 1.44, |
| "grad_norm": 18.75, |
| "learning_rate": 1.5562962962962965e-05, |
| "loss": 13.6293, |
| "mean_token_accuracy": 0.7877266079187393, |
| "num_tokens": 25606824.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_biology_entropy": 3.37851238822937, |
| "eval_biology_loss": 3.7537248134613037, |
| "eval_biology_mean_token_accuracy": 0.42950630354881286, |
| "eval_biology_num_tokens": 25606824.0, |
| "eval_biology_runtime": 49.7825, |
| "eval_biology_samples_per_second": 10.044, |
| "eval_biology_steps_per_second": 2.511, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_chemistry_entropy": 1.9617934045791625, |
| "eval_chemistry_loss": 2.017207145690918, |
| "eval_chemistry_mean_token_accuracy": 0.6149281630516052, |
| "eval_chemistry_num_tokens": 25606824.0, |
| "eval_chemistry_runtime": 61.6914, |
| "eval_chemistry_samples_per_second": 8.105, |
| "eval_chemistry_steps_per_second": 2.026, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_math_entropy": 0.8517674751281739, |
| "eval_math_loss": 0.8629322648048401, |
| "eval_math_mean_token_accuracy": 0.7848198075294495, |
| "eval_math_num_tokens": 25606824.0, |
| "eval_math_runtime": 63.2603, |
| "eval_math_samples_per_second": 7.904, |
| "eval_math_steps_per_second": 1.976, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_physics_entropy": 1.496031648159027, |
| "eval_physics_loss": 1.5447328090667725, |
| "eval_physics_mean_token_accuracy": 0.6830805406570435, |
| "eval_physics_num_tokens": 25606824.0, |
| "eval_physics_runtime": 72.0951, |
| "eval_physics_samples_per_second": 6.935, |
| "eval_physics_steps_per_second": 1.734, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.8286773880943656, |
| "epoch": 1.456, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.548888888888889e-05, |
| "loss": 13.1968, |
| "mean_token_accuracy": 0.7905922997742891, |
| "num_tokens": 25886396.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.8260000789538026, |
| "epoch": 1.472, |
| "grad_norm": 17.125, |
| "learning_rate": 1.5414814814814814e-05, |
| "loss": 13.2698, |
| "mean_token_accuracy": 0.790665191039443, |
| "num_tokens": 26163618.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.8521020160987973, |
| "epoch": 1.488, |
| "grad_norm": 19.0, |
| "learning_rate": 1.5340740740740744e-05, |
| "loss": 13.8606, |
| "mean_token_accuracy": 0.7840919081121683, |
| "num_tokens": 26438338.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.8122210435569286, |
| "epoch": 1.504, |
| "grad_norm": 21.0, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 13.1256, |
| "mean_token_accuracy": 0.7937585778534413, |
| "num_tokens": 26729255.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.8557528490200639, |
| "epoch": 1.52, |
| "grad_norm": 16.5, |
| "learning_rate": 1.5192592592592594e-05, |
| "loss": 13.6488, |
| "mean_token_accuracy": 0.7870340771973133, |
| "num_tokens": 27017935.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.8247261323034764, |
| "epoch": 1.536, |
| "grad_norm": 15.1875, |
| "learning_rate": 1.5118518518518519e-05, |
| "loss": 13.3195, |
| "mean_token_accuracy": 0.7908138386905194, |
| "num_tokens": 27306339.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.8346869014203548, |
| "epoch": 1.552, |
| "grad_norm": 17.375, |
| "learning_rate": 1.5044444444444445e-05, |
| "loss": 13.3502, |
| "mean_token_accuracy": 0.7901338282972574, |
| "num_tokens": 27591959.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.8050638379529118, |
| "epoch": 1.568, |
| "grad_norm": 15.625, |
| "learning_rate": 1.497037037037037e-05, |
| "loss": 13.0359, |
| "mean_token_accuracy": 0.7952109534293413, |
| "num_tokens": 27884398.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.8222229275852442, |
| "epoch": 1.584, |
| "grad_norm": 19.75, |
| "learning_rate": 1.4896296296296298e-05, |
| "loss": 13.2707, |
| "mean_token_accuracy": 0.7913329377770424, |
| "num_tokens": 28171274.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.8244392290711403, |
| "epoch": 1.6, |
| "grad_norm": 20.0, |
| "learning_rate": 1.4822222222222225e-05, |
| "loss": 13.1036, |
| "mean_token_accuracy": 0.7916798040270805, |
| "num_tokens": 28457624.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_biology_entropy": 3.3615621147155763, |
| "eval_biology_loss": 3.8202450275421143, |
| "eval_biology_mean_token_accuracy": 0.4290500741004944, |
| "eval_biology_num_tokens": 28457624.0, |
| "eval_biology_runtime": 49.8587, |
| "eval_biology_samples_per_second": 10.028, |
| "eval_biology_steps_per_second": 2.507, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_chemistry_entropy": 1.937631681919098, |
| "eval_chemistry_loss": 2.0187153816223145, |
| "eval_chemistry_mean_token_accuracy": 0.6154364368915558, |
| "eval_chemistry_num_tokens": 28457624.0, |
| "eval_chemistry_runtime": 61.3814, |
| "eval_chemistry_samples_per_second": 8.146, |
| "eval_chemistry_steps_per_second": 2.036, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_math_entropy": 0.8357908477783204, |
| "eval_math_loss": 0.857028603553772, |
| "eval_math_mean_token_accuracy": 0.7858444275856018, |
| "eval_math_num_tokens": 28457624.0, |
| "eval_math_runtime": 63.398, |
| "eval_math_samples_per_second": 7.887, |
| "eval_math_steps_per_second": 1.972, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_physics_entropy": 1.4734057002067567, |
| "eval_physics_loss": 1.5425677299499512, |
| "eval_physics_mean_token_accuracy": 0.6834757437705994, |
| "eval_physics_num_tokens": 28457624.0, |
| "eval_physics_runtime": 72.3807, |
| "eval_physics_samples_per_second": 6.908, |
| "eval_physics_steps_per_second": 1.727, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.8183613562956452, |
| "epoch": 1.616, |
| "grad_norm": 15.8125, |
| "learning_rate": 1.474814814814815e-05, |
| "loss": 13.3252, |
| "mean_token_accuracy": 0.789626444876194, |
| "num_tokens": 28743099.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.8398883065208793, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 18.25, |
| "learning_rate": 1.4674074074074076e-05, |
| "loss": 13.3764, |
| "mean_token_accuracy": 0.7887058384716511, |
| "num_tokens": 29017297.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.809584242105484, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 18.5, |
| "learning_rate": 1.46e-05, |
| "loss": 13.0665, |
| "mean_token_accuracy": 0.7953496225178242, |
| "num_tokens": 29303707.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.8161404246464372, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 18.0, |
| "learning_rate": 1.4525925925925927e-05, |
| "loss": 13.1101, |
| "mean_token_accuracy": 0.7914217729121447, |
| "num_tokens": 29597156.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.8350240230560303, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 16.375, |
| "learning_rate": 1.4451851851851852e-05, |
| "loss": 13.3906, |
| "mean_token_accuracy": 0.7913927461951971, |
| "num_tokens": 29883879.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.8350640160962939, |
| "epoch": 1.696, |
| "grad_norm": 18.75, |
| "learning_rate": 1.4377777777777779e-05, |
| "loss": 13.5113, |
| "mean_token_accuracy": 0.7893182434141636, |
| "num_tokens": 30165760.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.8064091168344021, |
| "epoch": 1.712, |
| "grad_norm": 17.0, |
| "learning_rate": 1.4303703703703703e-05, |
| "loss": 13.0678, |
| "mean_token_accuracy": 0.7943233713507653, |
| "num_tokens": 30460367.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.8139542754739523, |
| "epoch": 1.728, |
| "grad_norm": 23.25, |
| "learning_rate": 1.4229629629629632e-05, |
| "loss": 12.9895, |
| "mean_token_accuracy": 0.7952147454023362, |
| "num_tokens": 30739137.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.8204835077747703, |
| "epoch": 1.744, |
| "grad_norm": 18.5, |
| "learning_rate": 1.4155555555555556e-05, |
| "loss": 13.1414, |
| "mean_token_accuracy": 0.7930131111294031, |
| "num_tokens": 31022663.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.8065364792943001, |
| "epoch": 1.76, |
| "grad_norm": 17.375, |
| "learning_rate": 1.4081481481481483e-05, |
| "loss": 13.0088, |
| "mean_token_accuracy": 0.7931559670716524, |
| "num_tokens": 31306494.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_biology_entropy": 3.3287238235473633, |
| "eval_biology_loss": 3.7768468856811523, |
| "eval_biology_mean_token_accuracy": 0.4298539180755615, |
| "eval_biology_num_tokens": 31306494.0, |
| "eval_biology_runtime": 49.8201, |
| "eval_biology_samples_per_second": 10.036, |
| "eval_biology_steps_per_second": 2.509, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_chemistry_entropy": 1.9171943378448486, |
| "eval_chemistry_loss": 2.0048129558563232, |
| "eval_chemistry_mean_token_accuracy": 0.6168817098140716, |
| "eval_chemistry_num_tokens": 31306494.0, |
| "eval_chemistry_runtime": 61.7404, |
| "eval_chemistry_samples_per_second": 8.098, |
| "eval_chemistry_steps_per_second": 2.025, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_math_entropy": 0.8366622452735901, |
| "eval_math_loss": 0.851693332195282, |
| "eval_math_mean_token_accuracy": 0.7872344598770141, |
| "eval_math_num_tokens": 31306494.0, |
| "eval_math_runtime": 63.5475, |
| "eval_math_samples_per_second": 7.868, |
| "eval_math_steps_per_second": 1.967, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_physics_entropy": 1.4740583176612854, |
| "eval_physics_loss": 1.5396031141281128, |
| "eval_physics_mean_token_accuracy": 0.6840226426124573, |
| "eval_physics_num_tokens": 31306494.0, |
| "eval_physics_runtime": 72.1115, |
| "eval_physics_samples_per_second": 6.934, |
| "eval_physics_steps_per_second": 1.733, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.8262896530330182, |
| "epoch": 1.776, |
| "grad_norm": 17.75, |
| "learning_rate": 1.400740740740741e-05, |
| "loss": 13.274, |
| "mean_token_accuracy": 0.7924778901040554, |
| "num_tokens": 31595542.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.8307720847427845, |
| "epoch": 1.792, |
| "grad_norm": 17.125, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 13.4174, |
| "mean_token_accuracy": 0.7890391640365124, |
| "num_tokens": 31881189.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.8346689926460386, |
| "epoch": 1.808, |
| "grad_norm": 16.625, |
| "learning_rate": 1.385925925925926e-05, |
| "loss": 13.3517, |
| "mean_token_accuracy": 0.7908568613231182, |
| "num_tokens": 32164196.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.7776938879862427, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 18.0, |
| "learning_rate": 1.3785185185185186e-05, |
| "loss": 12.5276, |
| "mean_token_accuracy": 0.7996609631925822, |
| "num_tokens": 32441530.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.8081878338009119, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 19.75, |
| "learning_rate": 1.3711111111111112e-05, |
| "loss": 13.0399, |
| "mean_token_accuracy": 0.794286473467946, |
| "num_tokens": 32723145.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.8392048073932529, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 16.25, |
| "learning_rate": 1.3637037037037037e-05, |
| "loss": 13.4004, |
| "mean_token_accuracy": 0.7900693397969007, |
| "num_tokens": 33011263.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.8063056256622076, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 16.625, |
| "learning_rate": 1.3562962962962965e-05, |
| "loss": 13.0163, |
| "mean_token_accuracy": 0.7964250590652228, |
| "num_tokens": 33298921.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.8290531987324357, |
| "epoch": 1.888, |
| "grad_norm": 19.0, |
| "learning_rate": 1.3488888888888888e-05, |
| "loss": 13.2527, |
| "mean_token_accuracy": 0.7908387798815966, |
| "num_tokens": 33576243.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.8150693111121654, |
| "epoch": 1.904, |
| "grad_norm": 16.25, |
| "learning_rate": 1.3414814814814817e-05, |
| "loss": 13.1745, |
| "mean_token_accuracy": 0.7923198211938143, |
| "num_tokens": 33850968.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.7930012326687574, |
| "epoch": 1.92, |
| "grad_norm": 15.875, |
| "learning_rate": 1.3340740740740741e-05, |
| "loss": 12.7681, |
| "mean_token_accuracy": 0.7967522501945495, |
| "num_tokens": 34128558.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_biology_entropy": 3.286720132827759, |
| "eval_biology_loss": 3.7095859050750732, |
| "eval_biology_mean_token_accuracy": 0.43265608382225035, |
| "eval_biology_num_tokens": 34128558.0, |
| "eval_biology_runtime": 49.805, |
| "eval_biology_samples_per_second": 10.039, |
| "eval_biology_steps_per_second": 2.51, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_chemistry_entropy": 1.898814220905304, |
| "eval_chemistry_loss": 1.9991657733917236, |
| "eval_chemistry_mean_token_accuracy": 0.6174907529354096, |
| "eval_chemistry_num_tokens": 34128558.0, |
| "eval_chemistry_runtime": 61.7352, |
| "eval_chemistry_samples_per_second": 8.099, |
| "eval_chemistry_steps_per_second": 2.025, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_math_entropy": 0.8344658007621765, |
| "eval_math_loss": 0.8473447561264038, |
| "eval_math_mean_token_accuracy": 0.7876402115821839, |
| "eval_math_num_tokens": 34128558.0, |
| "eval_math_runtime": 63.3639, |
| "eval_math_samples_per_second": 7.891, |
| "eval_math_steps_per_second": 1.973, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_physics_entropy": 1.4668292608261109, |
| "eval_physics_loss": 1.5414491891860962, |
| "eval_physics_mean_token_accuracy": 0.6841273403167725, |
| "eval_physics_num_tokens": 34128558.0, |
| "eval_physics_runtime": 72.1478, |
| "eval_physics_samples_per_second": 6.93, |
| "eval_physics_steps_per_second": 1.733, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.822460556961596, |
| "epoch": 1.936, |
| "grad_norm": 22.375, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 13.1389, |
| "mean_token_accuracy": 0.7921592094004154, |
| "num_tokens": 34408056.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.807614759914577, |
| "epoch": 1.952, |
| "grad_norm": 19.625, |
| "learning_rate": 1.3192592592592594e-05, |
| "loss": 13.0525, |
| "mean_token_accuracy": 0.7933488007634878, |
| "num_tokens": 34684679.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.8354181025177241, |
| "epoch": 1.968, |
| "grad_norm": 18.375, |
| "learning_rate": 1.311851851851852e-05, |
| "loss": 13.5335, |
| "mean_token_accuracy": 0.788308460265398, |
| "num_tokens": 34971038.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.8086721919476986, |
| "epoch": 1.984, |
| "grad_norm": 16.5, |
| "learning_rate": 1.3044444444444446e-05, |
| "loss": 12.9467, |
| "mean_token_accuracy": 0.7951289299875498, |
| "num_tokens": 35262281.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.7930673930794001, |
| "epoch": 2.0, |
| "grad_norm": 22.75, |
| "learning_rate": 1.297037037037037e-05, |
| "loss": 12.7941, |
| "mean_token_accuracy": 0.79657237008214, |
| "num_tokens": 35560864.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.8225008610635995, |
| "epoch": 2.016, |
| "grad_norm": 15.25, |
| "learning_rate": 1.2896296296296299e-05, |
| "loss": 13.1411, |
| "mean_token_accuracy": 0.7936458811163902, |
| "num_tokens": 35846704.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.8110213376581669, |
| "epoch": 2.032, |
| "grad_norm": 19.125, |
| "learning_rate": 1.2822222222222222e-05, |
| "loss": 12.9766, |
| "mean_token_accuracy": 0.7942904278635978, |
| "num_tokens": 36128775.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.7852249689400196, |
| "epoch": 2.048, |
| "grad_norm": 18.0, |
| "learning_rate": 1.274814814814815e-05, |
| "loss": 12.5822, |
| "mean_token_accuracy": 0.7999643433839083, |
| "num_tokens": 36419504.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.782286673039198, |
| "epoch": 2.064, |
| "grad_norm": 17.25, |
| "learning_rate": 1.2674074074074075e-05, |
| "loss": 12.6666, |
| "mean_token_accuracy": 0.7982493124902248, |
| "num_tokens": 36706816.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.795855419524014, |
| "epoch": 2.08, |
| "grad_norm": 18.375, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 12.6569, |
| "mean_token_accuracy": 0.7973877970129252, |
| "num_tokens": 36988475.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_biology_entropy": 3.2089530782699587, |
| "eval_biology_loss": 3.7030863761901855, |
| "eval_biology_mean_token_accuracy": 0.43272937917709353, |
| "eval_biology_num_tokens": 36988475.0, |
| "eval_biology_runtime": 49.7903, |
| "eval_biology_samples_per_second": 10.042, |
| "eval_biology_steps_per_second": 2.511, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_chemistry_entropy": 1.8433975267410279, |
| "eval_chemistry_loss": 1.9999408721923828, |
| "eval_chemistry_mean_token_accuracy": 0.6180174436569214, |
| "eval_chemistry_num_tokens": 36988475.0, |
| "eval_chemistry_runtime": 61.7274, |
| "eval_chemistry_samples_per_second": 8.1, |
| "eval_chemistry_steps_per_second": 2.025, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_math_entropy": 0.8109442203044891, |
| "eval_math_loss": 0.8452479243278503, |
| "eval_math_mean_token_accuracy": 0.7881795573234558, |
| "eval_math_num_tokens": 36988475.0, |
| "eval_math_runtime": 63.3275, |
| "eval_math_samples_per_second": 7.895, |
| "eval_math_steps_per_second": 1.974, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_physics_entropy": 1.4158136811256408, |
| "eval_physics_loss": 1.5318591594696045, |
| "eval_physics_mean_token_accuracy": 0.6849393043518066, |
| "eval_physics_num_tokens": 36988475.0, |
| "eval_physics_runtime": 72.137, |
| "eval_physics_samples_per_second": 6.931, |
| "eval_physics_steps_per_second": 1.733, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.8043522318825126, |
| "epoch": 2.096, |
| "grad_norm": 17.375, |
| "learning_rate": 1.2525925925925928e-05, |
| "loss": 12.9984, |
| "mean_token_accuracy": 0.7961476005613803, |
| "num_tokens": 37270131.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.7804812841117382, |
| "epoch": 2.112, |
| "grad_norm": 16.375, |
| "learning_rate": 1.2451851851851853e-05, |
| "loss": 12.5609, |
| "mean_token_accuracy": 0.798238455876708, |
| "num_tokens": 37563537.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.7818844333291054, |
| "epoch": 2.128, |
| "grad_norm": 19.375, |
| "learning_rate": 1.237777777777778e-05, |
| "loss": 12.4773, |
| "mean_token_accuracy": 0.799476521462202, |
| "num_tokens": 37843959.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.7816478908061981, |
| "epoch": 2.144, |
| "grad_norm": 17.625, |
| "learning_rate": 1.2303703703703704e-05, |
| "loss": 12.6277, |
| "mean_token_accuracy": 0.7990053717046977, |
| "num_tokens": 38133092.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.7931677252054214, |
| "epoch": 2.16, |
| "grad_norm": 16.375, |
| "learning_rate": 1.222962962962963e-05, |
| "loss": 12.7116, |
| "mean_token_accuracy": 0.7987668544054032, |
| "num_tokens": 38421229.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.7619917875155806, |
| "epoch": 2.176, |
| "grad_norm": 19.75, |
| "learning_rate": 1.2155555555555555e-05, |
| "loss": 12.2236, |
| "mean_token_accuracy": 0.8035903237760067, |
| "num_tokens": 38708043.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.7887991681694985, |
| "epoch": 2.192, |
| "grad_norm": 16.875, |
| "learning_rate": 1.2081481481481484e-05, |
| "loss": 12.6936, |
| "mean_token_accuracy": 0.796902683749795, |
| "num_tokens": 38996930.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.7967741012573242, |
| "epoch": 2.208, |
| "grad_norm": 30.25, |
| "learning_rate": 1.2007407407407408e-05, |
| "loss": 12.9174, |
| "mean_token_accuracy": 0.7960829515010118, |
| "num_tokens": 39279481.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.7880135742947459, |
| "epoch": 2.224, |
| "grad_norm": 19.75, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 12.6788, |
| "mean_token_accuracy": 0.7990153830498457, |
| "num_tokens": 39569030.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.7916025001555681, |
| "epoch": 2.24, |
| "grad_norm": 18.75, |
| "learning_rate": 1.185925925925926e-05, |
| "loss": 12.7022, |
| "mean_token_accuracy": 0.7965882256627083, |
| "num_tokens": 39854873.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_biology_entropy": 3.280108278274536, |
| "eval_biology_loss": 3.7498559951782227, |
| "eval_biology_mean_token_accuracy": 0.43300582957267764, |
| "eval_biology_num_tokens": 39854873.0, |
| "eval_biology_runtime": 49.7856, |
| "eval_biology_samples_per_second": 10.043, |
| "eval_biology_steps_per_second": 2.511, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_chemistry_entropy": 1.8648150267601014, |
| "eval_chemistry_loss": 1.9987932443618774, |
| "eval_chemistry_mean_token_accuracy": 0.6183059096336365, |
| "eval_chemistry_num_tokens": 39854873.0, |
| "eval_chemistry_runtime": 61.751, |
| "eval_chemistry_samples_per_second": 8.097, |
| "eval_chemistry_steps_per_second": 2.024, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_math_entropy": 0.8120264797210693, |
| "eval_math_loss": 0.8429214954376221, |
| "eval_math_mean_token_accuracy": 0.7886019821166992, |
| "eval_math_num_tokens": 39854873.0, |
| "eval_math_runtime": 63.3235, |
| "eval_math_samples_per_second": 7.896, |
| "eval_math_steps_per_second": 1.974, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_physics_entropy": 1.4317502527236938, |
| "eval_physics_loss": 1.5286813974380493, |
| "eval_physics_mean_token_accuracy": 0.6863654246330261, |
| "eval_physics_num_tokens": 39854873.0, |
| "eval_physics_runtime": 72.133, |
| "eval_physics_samples_per_second": 6.932, |
| "eval_physics_steps_per_second": 1.733, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.7847368892282247, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 17.25, |
| "learning_rate": 1.1785185185185186e-05, |
| "loss": 12.6258, |
| "mean_token_accuracy": 0.7981156852096319, |
| "num_tokens": 40141190.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.8105148211121559, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 22.75, |
| "learning_rate": 1.1711111111111113e-05, |
| "loss": 13.0268, |
| "mean_token_accuracy": 0.7950900815427303, |
| "num_tokens": 40415203.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.7900150781497359, |
| "epoch": 2.288, |
| "grad_norm": 17.5, |
| "learning_rate": 1.1637037037037037e-05, |
| "loss": 12.6047, |
| "mean_token_accuracy": 0.7972030244767666, |
| "num_tokens": 40702393.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.7867585562169552, |
| "epoch": 2.304, |
| "grad_norm": 21.125, |
| "learning_rate": 1.1562962962962964e-05, |
| "loss": 12.6491, |
| "mean_token_accuracy": 0.7997174769639969, |
| "num_tokens": 40982775.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.8002804208546876, |
| "epoch": 2.32, |
| "grad_norm": 20.875, |
| "learning_rate": 1.1488888888888889e-05, |
| "loss": 12.8551, |
| "mean_token_accuracy": 0.7973973777145147, |
| "num_tokens": 41263356.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.7985019231215119, |
| "epoch": 2.336, |
| "grad_norm": 18.125, |
| "learning_rate": 1.1414814814814817e-05, |
| "loss": 12.8916, |
| "mean_token_accuracy": 0.7960489392280579, |
| "num_tokens": 41545235.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.8076209686696529, |
| "epoch": 2.352, |
| "grad_norm": 18.75, |
| "learning_rate": 1.1340740740740742e-05, |
| "loss": 12.9997, |
| "mean_token_accuracy": 0.7936360351741314, |
| "num_tokens": 41833417.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.7910049550235272, |
| "epoch": 2.368, |
| "grad_norm": 19.0, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 12.7404, |
| "mean_token_accuracy": 0.7980276115238667, |
| "num_tokens": 42117030.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.811725552752614, |
| "epoch": 2.384, |
| "grad_norm": 22.875, |
| "learning_rate": 1.1192592592592593e-05, |
| "loss": 13.0613, |
| "mean_token_accuracy": 0.793327846005559, |
| "num_tokens": 42410990.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.8014231506735087, |
| "epoch": 2.4, |
| "grad_norm": 20.0, |
| "learning_rate": 1.111851851851852e-05, |
| "loss": 12.8548, |
| "mean_token_accuracy": 0.796426559984684, |
| "num_tokens": 42691890.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_biology_entropy": 3.213496032714844, |
| "eval_biology_loss": 3.7113101482391357, |
| "eval_biology_mean_token_accuracy": 0.43321818923950195, |
| "eval_biology_num_tokens": 42691890.0, |
| "eval_biology_runtime": 49.9484, |
| "eval_biology_samples_per_second": 10.01, |
| "eval_biology_steps_per_second": 2.503, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_chemistry_entropy": 1.8208427929878235, |
| "eval_chemistry_loss": 1.9899694919586182, |
| "eval_chemistry_mean_token_accuracy": 0.6187239620685577, |
| "eval_chemistry_num_tokens": 42691890.0, |
| "eval_chemistry_runtime": 61.2857, |
| "eval_chemistry_samples_per_second": 8.159, |
| "eval_chemistry_steps_per_second": 2.04, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_math_entropy": 0.7984081234931946, |
| "eval_math_loss": 0.8405261635780334, |
| "eval_math_mean_token_accuracy": 0.7889668345451355, |
| "eval_math_num_tokens": 42691890.0, |
| "eval_math_runtime": 63.266, |
| "eval_math_samples_per_second": 7.903, |
| "eval_math_steps_per_second": 1.976, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_physics_entropy": 1.4056278495788574, |
| "eval_physics_loss": 1.5309016704559326, |
| "eval_physics_mean_token_accuracy": 0.6855477833747864, |
| "eval_physics_num_tokens": 42691890.0, |
| "eval_physics_runtime": 72.1465, |
| "eval_physics_samples_per_second": 6.93, |
| "eval_physics_steps_per_second": 1.733, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.8084384858608246, |
| "epoch": 2.416, |
| "grad_norm": 20.5, |
| "learning_rate": 1.1044444444444444e-05, |
| "loss": 13.0963, |
| "mean_token_accuracy": 0.792670714110136, |
| "num_tokens": 42971588.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.7932598620653153, |
| "epoch": 2.432, |
| "grad_norm": 19.875, |
| "learning_rate": 1.0970370370370371e-05, |
| "loss": 12.668, |
| "mean_token_accuracy": 0.79843137152493, |
| "num_tokens": 43253821.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.7650916231796145, |
| "epoch": 2.448, |
| "grad_norm": 18.5, |
| "learning_rate": 1.0896296296296298e-05, |
| "loss": 12.4535, |
| "mean_token_accuracy": 0.8016320440918208, |
| "num_tokens": 43550902.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.8061025515198708, |
| "epoch": 2.464, |
| "grad_norm": 18.125, |
| "learning_rate": 1.0822222222222222e-05, |
| "loss": 12.8029, |
| "mean_token_accuracy": 0.7956046860665082, |
| "num_tokens": 43844259.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.8425655115395785, |
| "epoch": 2.48, |
| "grad_norm": 31.125, |
| "learning_rate": 1.074814814814815e-05, |
| "loss": 13.5112, |
| "mean_token_accuracy": 0.7877779383212328, |
| "num_tokens": 44115701.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.8077523436397314, |
| "epoch": 2.496, |
| "grad_norm": 17.375, |
| "learning_rate": 1.0674074074074074e-05, |
| "loss": 13.0914, |
| "mean_token_accuracy": 0.7937187146395445, |
| "num_tokens": 44405520.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.7986883038654924, |
| "epoch": 2.512, |
| "grad_norm": 17.0, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 12.7733, |
| "mean_token_accuracy": 0.7968008328229189, |
| "num_tokens": 44686477.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.7973098121583462, |
| "epoch": 2.528, |
| "grad_norm": 17.625, |
| "learning_rate": 1.0525925925925927e-05, |
| "loss": 12.8409, |
| "mean_token_accuracy": 0.7960524547845125, |
| "num_tokens": 44969760.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.793004597350955, |
| "epoch": 2.544, |
| "grad_norm": 18.625, |
| "learning_rate": 1.0451851851851853e-05, |
| "loss": 12.7374, |
| "mean_token_accuracy": 0.7978652317076922, |
| "num_tokens": 45255326.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.7932099934667349, |
| "epoch": 2.56, |
| "grad_norm": 17.25, |
| "learning_rate": 1.0377777777777778e-05, |
| "loss": 12.7191, |
| "mean_token_accuracy": 0.7962283626198768, |
| "num_tokens": 45532525.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_biology_entropy": 3.2518484830856322, |
| "eval_biology_loss": 3.700951099395752, |
| "eval_biology_mean_token_accuracy": 0.4338484408855438, |
| "eval_biology_num_tokens": 45532525.0, |
| "eval_biology_runtime": 49.825, |
| "eval_biology_samples_per_second": 10.035, |
| "eval_biology_steps_per_second": 2.509, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_chemistry_entropy": 1.8449390592575072, |
| "eval_chemistry_loss": 1.9900366067886353, |
| "eval_chemistry_mean_token_accuracy": 0.6188704285621643, |
| "eval_chemistry_num_tokens": 45532525.0, |
| "eval_chemistry_runtime": 61.7822, |
| "eval_chemistry_samples_per_second": 8.093, |
| "eval_chemistry_steps_per_second": 2.023, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_math_entropy": 0.8004411847591401, |
| "eval_math_loss": 0.8377605080604553, |
| "eval_math_mean_token_accuracy": 0.7894859976768493, |
| "eval_math_num_tokens": 45532525.0, |
| "eval_math_runtime": 63.3799, |
| "eval_math_samples_per_second": 7.889, |
| "eval_math_steps_per_second": 1.972, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_physics_entropy": 1.4191389164924622, |
| "eval_physics_loss": 1.5280442237854004, |
| "eval_physics_mean_token_accuracy": 0.6858750200271606, |
| "eval_physics_num_tokens": 45532525.0, |
| "eval_physics_runtime": 72.1737, |
| "eval_physics_samples_per_second": 6.928, |
| "eval_physics_steps_per_second": 1.732, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.7856329085305334, |
| "epoch": 2.576, |
| "grad_norm": 18.25, |
| "learning_rate": 1.0303703703703705e-05, |
| "loss": 12.7499, |
| "mean_token_accuracy": 0.7975605644285679, |
| "num_tokens": 45817478.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.7819217784330249, |
| "epoch": 2.592, |
| "grad_norm": 31.375, |
| "learning_rate": 1.0229629629629631e-05, |
| "loss": 12.5615, |
| "mean_token_accuracy": 0.7989848144352436, |
| "num_tokens": 46109575.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.8088566403836012, |
| "epoch": 2.608, |
| "grad_norm": 21.375, |
| "learning_rate": 1.0155555555555556e-05, |
| "loss": 13.044, |
| "mean_token_accuracy": 0.7929384857416153, |
| "num_tokens": 46391461.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.7695744272321463, |
| "epoch": 2.624, |
| "grad_norm": 18.0, |
| "learning_rate": 1.0081481481481484e-05, |
| "loss": 12.3887, |
| "mean_token_accuracy": 0.8046348694711923, |
| "num_tokens": 46683117.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.804672497138381, |
| "epoch": 2.64, |
| "grad_norm": 18.625, |
| "learning_rate": 1.0007407407407407e-05, |
| "loss": 12.8594, |
| "mean_token_accuracy": 0.7948314756155014, |
| "num_tokens": 46965534.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.7649372734129429, |
| "epoch": 2.656, |
| "grad_norm": 18.125, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 12.2923, |
| "mean_token_accuracy": 0.8014582961797714, |
| "num_tokens": 47245852.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.8046383403241635, |
| "epoch": 2.672, |
| "grad_norm": 22.0, |
| "learning_rate": 9.85925925925926e-06, |
| "loss": 12.9533, |
| "mean_token_accuracy": 0.7926652491092682, |
| "num_tokens": 47524916.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.7880237903445959, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 18.75, |
| "learning_rate": 9.785185185185187e-06, |
| "loss": 12.7005, |
| "mean_token_accuracy": 0.7974669005721807, |
| "num_tokens": 47807131.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.8009384609758854, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 21.125, |
| "learning_rate": 9.711111111111111e-06, |
| "loss": 12.8682, |
| "mean_token_accuracy": 0.7949213892221451, |
| "num_tokens": 48099654.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.7890177585184575, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 19.125, |
| "learning_rate": 9.637037037037038e-06, |
| "loss": 12.702, |
| "mean_token_accuracy": 0.7969291642308235, |
| "num_tokens": 48375019.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_biology_entropy": 3.2781650161743165, |
| "eval_biology_loss": 3.7301559448242188, |
| "eval_biology_mean_token_accuracy": 0.43347722268104555, |
| "eval_biology_num_tokens": 48375019.0, |
| "eval_biology_runtime": 49.7998, |
| "eval_biology_samples_per_second": 10.04, |
| "eval_biology_steps_per_second": 2.51, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_chemistry_entropy": 1.8482142910957335, |
| "eval_chemistry_loss": 1.9887773990631104, |
| "eval_chemistry_mean_token_accuracy": 0.6194968128204346, |
| "eval_chemistry_num_tokens": 48375019.0, |
| "eval_chemistry_runtime": 62.2368, |
| "eval_chemistry_samples_per_second": 8.034, |
| "eval_chemistry_steps_per_second": 2.008, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_math_entropy": 0.809392058134079, |
| "eval_math_loss": 0.8359456062316895, |
| "eval_math_mean_token_accuracy": 0.7895615015029908, |
| "eval_math_num_tokens": 48375019.0, |
| "eval_math_runtime": 63.3325, |
| "eval_math_samples_per_second": 7.895, |
| "eval_math_steps_per_second": 1.974, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_physics_entropy": 1.421650909423828, |
| "eval_physics_loss": 1.5206794738769531, |
| "eval_physics_mean_token_accuracy": 0.6867666440010071, |
| "eval_physics_num_tokens": 48375019.0, |
| "eval_physics_runtime": 72.209, |
| "eval_physics_samples_per_second": 6.924, |
| "eval_physics_steps_per_second": 1.731, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.7974690224975347, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 19.75, |
| "learning_rate": 9.562962962962965e-06, |
| "loss": 12.7825, |
| "mean_token_accuracy": 0.7953499432653188, |
| "num_tokens": 48659284.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.7819192102178931, |
| "epoch": 2.752, |
| "grad_norm": 17.75, |
| "learning_rate": 9.48888888888889e-06, |
| "loss": 12.6506, |
| "mean_token_accuracy": 0.7982836693525315, |
| "num_tokens": 48943804.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.7997678998857737, |
| "epoch": 2.768, |
| "grad_norm": 18.625, |
| "learning_rate": 9.414814814814816e-06, |
| "loss": 12.7813, |
| "mean_token_accuracy": 0.7972954675555229, |
| "num_tokens": 49230939.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.8081969961524009, |
| "epoch": 2.784, |
| "grad_norm": 19.75, |
| "learning_rate": 9.34074074074074e-06, |
| "loss": 13.0832, |
| "mean_token_accuracy": 0.7918697893619537, |
| "num_tokens": 49504425.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.7892359040677548, |
| "epoch": 2.8, |
| "grad_norm": 15.375, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 12.7791, |
| "mean_token_accuracy": 0.798084444925189, |
| "num_tokens": 49782661.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.7996826503425837, |
| "epoch": 2.816, |
| "grad_norm": 22.75, |
| "learning_rate": 9.192592592592594e-06, |
| "loss": 12.8429, |
| "mean_token_accuracy": 0.7955689582973718, |
| "num_tokens": 50073632.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.7853869337588548, |
| "epoch": 2.832, |
| "grad_norm": 19.625, |
| "learning_rate": 9.118518518518518e-06, |
| "loss": 12.607, |
| "mean_token_accuracy": 0.7983032051473856, |
| "num_tokens": 50356964.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.8069319518283009, |
| "epoch": 2.848, |
| "grad_norm": 16.25, |
| "learning_rate": 9.044444444444445e-06, |
| "loss": 13.0585, |
| "mean_token_accuracy": 0.7950124684721231, |
| "num_tokens": 50644535.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.8049573350697756, |
| "epoch": 2.864, |
| "grad_norm": 21.25, |
| "learning_rate": 8.970370370370372e-06, |
| "loss": 12.8888, |
| "mean_token_accuracy": 0.7958490017801523, |
| "num_tokens": 50925653.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.8024192716926336, |
| "epoch": 2.88, |
| "grad_norm": 19.375, |
| "learning_rate": 8.896296296296298e-06, |
| "loss": 12.7601, |
| "mean_token_accuracy": 0.7968584690243006, |
| "num_tokens": 51204374.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_biology_entropy": 3.233502674102783, |
| "eval_biology_loss": 3.723733425140381, |
| "eval_biology_mean_token_accuracy": 0.43377703070640566, |
| "eval_biology_num_tokens": 51204374.0, |
| "eval_biology_runtime": 50.4406, |
| "eval_biology_samples_per_second": 9.913, |
| "eval_biology_steps_per_second": 2.478, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_chemistry_entropy": 1.8262592434883118, |
| "eval_chemistry_loss": 1.9915226697921753, |
| "eval_chemistry_mean_token_accuracy": 0.6196319291591644, |
| "eval_chemistry_num_tokens": 51204374.0, |
| "eval_chemistry_runtime": 62.4404, |
| "eval_chemistry_samples_per_second": 8.008, |
| "eval_chemistry_steps_per_second": 2.002, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_math_entropy": 0.801637535572052, |
| "eval_math_loss": 0.8344833850860596, |
| "eval_math_mean_token_accuracy": 0.7902569670677185, |
| "eval_math_num_tokens": 51204374.0, |
| "eval_math_runtime": 63.6911, |
| "eval_math_samples_per_second": 7.85, |
| "eval_math_steps_per_second": 1.963, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_physics_entropy": 1.4114351081848144, |
| "eval_physics_loss": 1.5217763185501099, |
| "eval_physics_mean_token_accuracy": 0.6866876316070557, |
| "eval_physics_num_tokens": 51204374.0, |
| "eval_physics_runtime": 72.2495, |
| "eval_physics_samples_per_second": 6.92, |
| "eval_physics_steps_per_second": 1.73, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.7727437706664204, |
| "epoch": 2.896, |
| "grad_norm": 20.5, |
| "learning_rate": 8.822222222222223e-06, |
| "loss": 12.5378, |
| "mean_token_accuracy": 0.7991547420620918, |
| "num_tokens": 51483944.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.7949327051639556, |
| "epoch": 2.912, |
| "grad_norm": 16.875, |
| "learning_rate": 8.74814814814815e-06, |
| "loss": 12.8129, |
| "mean_token_accuracy": 0.7974283535033464, |
| "num_tokens": 51765755.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.7891913030296565, |
| "epoch": 2.928, |
| "grad_norm": 19.875, |
| "learning_rate": 8.674074074074074e-06, |
| "loss": 12.7001, |
| "mean_token_accuracy": 0.7998170506209135, |
| "num_tokens": 52056379.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.7793466189876199, |
| "epoch": 2.944, |
| "grad_norm": 16.75, |
| "learning_rate": 8.6e-06, |
| "loss": 12.4224, |
| "mean_token_accuracy": 0.7998687323182821, |
| "num_tokens": 52346232.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.7924864767119288, |
| "epoch": 2.96, |
| "grad_norm": 17.0, |
| "learning_rate": 8.525925925925927e-06, |
| "loss": 12.7675, |
| "mean_token_accuracy": 0.7973909642547369, |
| "num_tokens": 52633789.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.7772072922438383, |
| "epoch": 2.976, |
| "grad_norm": 16.625, |
| "learning_rate": 8.451851851851852e-06, |
| "loss": 12.5696, |
| "mean_token_accuracy": 0.798350129276514, |
| "num_tokens": 52911755.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.785734941251576, |
| "epoch": 2.992, |
| "grad_norm": 18.75, |
| "learning_rate": 8.377777777777779e-06, |
| "loss": 12.5958, |
| "mean_token_accuracy": 0.798584484308958, |
| "num_tokens": 53198176.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.798486184887588, |
| "epoch": 3.008, |
| "grad_norm": 17.625, |
| "learning_rate": 8.303703703703705e-06, |
| "loss": 12.8286, |
| "mean_token_accuracy": 0.7952684313058853, |
| "num_tokens": 53481893.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.798673938214779, |
| "epoch": 3.024, |
| "grad_norm": 17.375, |
| "learning_rate": 8.229629629629632e-06, |
| "loss": 12.7458, |
| "mean_token_accuracy": 0.7966153182089328, |
| "num_tokens": 53771717.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.7936699235811829, |
| "epoch": 3.04, |
| "grad_norm": 18.125, |
| "learning_rate": 8.155555555555556e-06, |
| "loss": 12.7437, |
| "mean_token_accuracy": 0.7971193008124828, |
| "num_tokens": 54058045.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_biology_entropy": 3.2133601503372193, |
| "eval_biology_loss": 3.722015142440796, |
| "eval_biology_mean_token_accuracy": 0.43406607151031495, |
| "eval_biology_num_tokens": 54058045.0, |
| "eval_biology_runtime": 50.1009, |
| "eval_biology_samples_per_second": 9.98, |
| "eval_biology_steps_per_second": 2.495, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_chemistry_entropy": 1.8122160468101502, |
| "eval_chemistry_loss": 1.9867562055587769, |
| "eval_chemistry_mean_token_accuracy": 0.6197145907878876, |
| "eval_chemistry_num_tokens": 54058045.0, |
| "eval_chemistry_runtime": 62.5867, |
| "eval_chemistry_samples_per_second": 7.989, |
| "eval_chemistry_steps_per_second": 1.997, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_math_entropy": 0.7948782880306244, |
| "eval_math_loss": 0.8342062830924988, |
| "eval_math_mean_token_accuracy": 0.7903887577056885, |
| "eval_math_num_tokens": 54058045.0, |
| "eval_math_runtime": 64.4647, |
| "eval_math_samples_per_second": 7.756, |
| "eval_math_steps_per_second": 1.939, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_physics_entropy": 1.4016832985877992, |
| "eval_physics_loss": 1.5218627452850342, |
| "eval_physics_mean_token_accuracy": 0.6867203059196472, |
| "eval_physics_num_tokens": 54058045.0, |
| "eval_physics_runtime": 72.1233, |
| "eval_physics_samples_per_second": 6.933, |
| "eval_physics_steps_per_second": 1.733, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.791794583760202, |
| "epoch": 3.056, |
| "grad_norm": 16.0, |
| "learning_rate": 8.081481481481483e-06, |
| "loss": 12.767, |
| "mean_token_accuracy": 0.7971944514662027, |
| "num_tokens": 54334332.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.7716375244781375, |
| "epoch": 3.072, |
| "grad_norm": 17.125, |
| "learning_rate": 8.007407407407408e-06, |
| "loss": 12.435, |
| "mean_token_accuracy": 0.8003965705633164, |
| "num_tokens": 54624543.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.7641144976019859, |
| "epoch": 3.088, |
| "grad_norm": 17.875, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 12.2604, |
| "mean_token_accuracy": 0.8041298471391201, |
| "num_tokens": 54907550.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.7768820196390152, |
| "epoch": 3.104, |
| "grad_norm": 15.3125, |
| "learning_rate": 7.859259259259259e-06, |
| "loss": 12.4156, |
| "mean_token_accuracy": 0.8015246070921421, |
| "num_tokens": 55190959.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.767643989995122, |
| "epoch": 3.12, |
| "grad_norm": 19.0, |
| "learning_rate": 7.785185185185185e-06, |
| "loss": 12.2748, |
| "mean_token_accuracy": 0.8037003818899393, |
| "num_tokens": 55481635.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.7627615902572871, |
| "epoch": 3.136, |
| "grad_norm": 17.5, |
| "learning_rate": 7.711111111111112e-06, |
| "loss": 12.3626, |
| "mean_token_accuracy": 0.8015805222094059, |
| "num_tokens": 55769010.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.7825732477009296, |
| "epoch": 3.152, |
| "grad_norm": 18.0, |
| "learning_rate": 7.637037037037037e-06, |
| "loss": 12.5732, |
| "mean_token_accuracy": 0.7999679546803236, |
| "num_tokens": 56053160.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.7954605100676417, |
| "epoch": 3.168, |
| "grad_norm": 16.875, |
| "learning_rate": 7.562962962962963e-06, |
| "loss": 12.7728, |
| "mean_token_accuracy": 0.7977348294109106, |
| "num_tokens": 56337066.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.7740211399272084, |
| "epoch": 3.184, |
| "grad_norm": 19.625, |
| "learning_rate": 7.48888888888889e-06, |
| "loss": 12.3752, |
| "mean_token_accuracy": 0.799398972466588, |
| "num_tokens": 56618899.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.7562787082046271, |
| "epoch": 3.2, |
| "grad_norm": 14.875, |
| "learning_rate": 7.4148148148148155e-06, |
| "loss": 12.1475, |
| "mean_token_accuracy": 0.8042113542556762, |
| "num_tokens": 56910071.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_biology_entropy": 3.208591146469116, |
| "eval_biology_loss": 3.7341694831848145, |
| "eval_biology_mean_token_accuracy": 0.4337711172103882, |
| "eval_biology_num_tokens": 56910071.0, |
| "eval_biology_runtime": 49.9923, |
| "eval_biology_samples_per_second": 10.002, |
| "eval_biology_steps_per_second": 2.5, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_chemistry_entropy": 1.8099074120521546, |
| "eval_chemistry_loss": 1.998437762260437, |
| "eval_chemistry_mean_token_accuracy": 0.6191729652881622, |
| "eval_chemistry_num_tokens": 56910071.0, |
| "eval_chemistry_runtime": 61.2788, |
| "eval_chemistry_samples_per_second": 8.159, |
| "eval_chemistry_steps_per_second": 2.04, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_math_entropy": 0.7919997727870941, |
| "eval_math_loss": 0.8333685398101807, |
| "eval_math_mean_token_accuracy": 0.7902761344909668, |
| "eval_math_num_tokens": 56910071.0, |
| "eval_math_runtime": 63.3474, |
| "eval_math_samples_per_second": 7.893, |
| "eval_math_steps_per_second": 1.973, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_physics_entropy": 1.3938024516105652, |
| "eval_physics_loss": 1.5245832204818726, |
| "eval_physics_mean_token_accuracy": 0.6867119417190551, |
| "eval_physics_num_tokens": 56910071.0, |
| "eval_physics_runtime": 72.2962, |
| "eval_physics_samples_per_second": 6.916, |
| "eval_physics_steps_per_second": 1.729, |
| "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 |
| } |
|
|
