| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.8, |
| "eval_steps": 100, |
| "global_step": 500, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.5395457262173295, |
| "epoch": 0.016, |
| "grad_norm": 3200.0, |
| "learning_rate": 4.5000000000000003e-07, |
| "loss": 92.5745, |
| "mean_token_accuracy": 0.4849457062780857, |
| "num_tokens": 195524.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.5515131525695324, |
| "epoch": 0.032, |
| "grad_norm": 2912.0, |
| "learning_rate": 9.500000000000001e-07, |
| "loss": 88.931, |
| "mean_token_accuracy": 0.4874983370304108, |
| "num_tokens": 390903.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.6169477274641395, |
| "epoch": 0.048, |
| "grad_norm": 2912.0, |
| "learning_rate": 1.45e-06, |
| "loss": 78.3695, |
| "mean_token_accuracy": 0.4966813003644347, |
| "num_tokens": 589868.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.8348326332867145, |
| "epoch": 0.064, |
| "grad_norm": 1160.0, |
| "learning_rate": 1.9500000000000004e-06, |
| "loss": 60.9248, |
| "mean_token_accuracy": 0.505222667939961, |
| "num_tokens": 791190.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.3063700653612613, |
| "epoch": 0.08, |
| "grad_norm": 247.0, |
| "learning_rate": 2.4500000000000003e-06, |
| "loss": 46.7906, |
| "mean_token_accuracy": 0.5185849336907268, |
| "num_tokens": 989860.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.959995786100626, |
| "epoch": 0.096, |
| "grad_norm": 81.0, |
| "learning_rate": 2.95e-06, |
| "loss": 39.2759, |
| "mean_token_accuracy": 0.5406044816598297, |
| "num_tokens": 1181777.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 2.191839651763439, |
| "epoch": 0.112, |
| "grad_norm": 57.0, |
| "learning_rate": 3.45e-06, |
| "loss": 34.17, |
| "mean_token_accuracy": 0.5669999368488788, |
| "num_tokens": 1385513.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.863330227136612, |
| "epoch": 0.128, |
| "grad_norm": 48.25, |
| "learning_rate": 3.95e-06, |
| "loss": 31.3942, |
| "mean_token_accuracy": 0.5891917455941439, |
| "num_tokens": 1582368.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.8903757609426974, |
| "epoch": 0.144, |
| "grad_norm": 39.25, |
| "learning_rate": 4.450000000000001e-06, |
| "loss": 29.9275, |
| "mean_token_accuracy": 0.6045342192053795, |
| "num_tokens": 1773764.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.8238266378641128, |
| "epoch": 0.16, |
| "grad_norm": 36.25, |
| "learning_rate": 4.95e-06, |
| "loss": 29.3315, |
| "mean_token_accuracy": 0.6066708620637655, |
| "num_tokens": 1970077.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 1.7597194662094116, |
| "eval_biology_loss": 1.708297848701477, |
| "eval_biology_mean_token_accuracy": 0.6260299117565155, |
| "eval_biology_num_tokens": 1970077.0, |
| "eval_biology_runtime": 36.35, |
| "eval_biology_samples_per_second": 13.755, |
| "eval_biology_steps_per_second": 3.439, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 1.278447434425354, |
| "eval_chemistry_loss": 1.24350905418396, |
| "eval_chemistry_mean_token_accuracy": 0.717205593585968, |
| "eval_chemistry_num_tokens": 1970077.0, |
| "eval_chemistry_runtime": 44.9852, |
| "eval_chemistry_samples_per_second": 11.115, |
| "eval_chemistry_steps_per_second": 2.779, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 0.9187946987152099, |
| "eval_math_loss": 1.206222414970398, |
| "eval_math_mean_token_accuracy": 0.7349825048446655, |
| "eval_math_num_tokens": 1970077.0, |
| "eval_math_runtime": 46.1658, |
| "eval_math_samples_per_second": 10.831, |
| "eval_math_steps_per_second": 2.708, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 1.2150983595848084, |
| "eval_physics_loss": 1.2188719511032104, |
| "eval_physics_mean_token_accuracy": 0.727249002456665, |
| "eval_physics_num_tokens": 1970077.0, |
| "eval_physics_runtime": 52.9985, |
| "eval_physics_samples_per_second": 9.434, |
| "eval_physics_steps_per_second": 2.359, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.7616284050047397, |
| "epoch": 0.176, |
| "grad_norm": 33.25, |
| "learning_rate": 5.450000000000001e-06, |
| "loss": 27.9906, |
| "mean_token_accuracy": 0.6186339979991317, |
| "num_tokens": 2168354.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.7145348988473414, |
| "epoch": 0.192, |
| "grad_norm": 36.25, |
| "learning_rate": 5.950000000000001e-06, |
| "loss": 27.271, |
| "mean_token_accuracy": 0.6236628353595733, |
| "num_tokens": 2365822.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.7000788144767285, |
| "epoch": 0.208, |
| "grad_norm": 34.75, |
| "learning_rate": 6.450000000000001e-06, |
| "loss": 26.7483, |
| "mean_token_accuracy": 0.6299795724451542, |
| "num_tokens": 2558762.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.6474159061908722, |
| "epoch": 0.224, |
| "grad_norm": 30.875, |
| "learning_rate": 6.95e-06, |
| "loss": 25.9891, |
| "mean_token_accuracy": 0.6365062229335308, |
| "num_tokens": 2755347.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.6314032576978206, |
| "epoch": 0.24, |
| "grad_norm": 31.125, |
| "learning_rate": 7.450000000000001e-06, |
| "loss": 25.1799, |
| "mean_token_accuracy": 0.64104891307652, |
| "num_tokens": 2947346.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.609092117100954, |
| "epoch": 0.256, |
| "grad_norm": 28.5, |
| "learning_rate": 7.950000000000002e-06, |
| "loss": 24.976, |
| "mean_token_accuracy": 0.6415446903556585, |
| "num_tokens": 3139957.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.5253454238176345, |
| "epoch": 0.272, |
| "grad_norm": 28.875, |
| "learning_rate": 8.45e-06, |
| "loss": 24.4431, |
| "mean_token_accuracy": 0.6438875071704387, |
| "num_tokens": 3335951.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.5003480702638625, |
| "epoch": 0.288, |
| "grad_norm": 31.5, |
| "learning_rate": 8.95e-06, |
| "loss": 23.9226, |
| "mean_token_accuracy": 0.6480991121381521, |
| "num_tokens": 3539731.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.4835937578231095, |
| "epoch": 0.304, |
| "grad_norm": 29.375, |
| "learning_rate": 9.450000000000001e-06, |
| "loss": 23.7553, |
| "mean_token_accuracy": 0.6533681858330965, |
| "num_tokens": 3733488.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.501522469520569, |
| "epoch": 0.32, |
| "grad_norm": 31.625, |
| "learning_rate": 9.950000000000001e-06, |
| "loss": 24.0263, |
| "mean_token_accuracy": 0.6483922835439444, |
| "num_tokens": 3920545.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 1.430639804840088, |
| "eval_biology_loss": 1.4240669012069702, |
| "eval_biology_mean_token_accuracy": 0.6643721928596497, |
| "eval_biology_num_tokens": 3920545.0, |
| "eval_biology_runtime": 36.631, |
| "eval_biology_samples_per_second": 13.65, |
| "eval_biology_steps_per_second": 3.412, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 1.068233790397644, |
| "eval_chemistry_loss": 1.0446134805679321, |
| "eval_chemistry_mean_token_accuracy": 0.7419922127723694, |
| "eval_chemistry_num_tokens": 3920545.0, |
| "eval_chemistry_runtime": 45.0089, |
| "eval_chemistry_samples_per_second": 11.109, |
| "eval_chemistry_steps_per_second": 2.777, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 0.8016578803062439, |
| "eval_math_loss": 1.064239740371704, |
| "eval_math_mean_token_accuracy": 0.7532512550354004, |
| "eval_math_num_tokens": 3920545.0, |
| "eval_math_runtime": 46.1998, |
| "eval_math_samples_per_second": 10.823, |
| "eval_math_steps_per_second": 2.706, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 1.0390429463386535, |
| "eval_physics_loss": 1.033566951751709, |
| "eval_physics_mean_token_accuracy": 0.751745885848999, |
| "eval_physics_num_tokens": 3920545.0, |
| "eval_physics_runtime": 53.0382, |
| "eval_physics_samples_per_second": 9.427, |
| "eval_physics_steps_per_second": 2.357, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.4409266009926796, |
| "epoch": 0.336, |
| "grad_norm": 30.5, |
| "learning_rate": 1.045e-05, |
| "loss": 23.1469, |
| "mean_token_accuracy": 0.6563692390918732, |
| "num_tokens": 4114077.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.4634413696825503, |
| "epoch": 0.352, |
| "grad_norm": 29.875, |
| "learning_rate": 1.095e-05, |
| "loss": 23.4319, |
| "mean_token_accuracy": 0.6559967070817947, |
| "num_tokens": 4306949.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 1.4390382423996926, |
| "epoch": 0.368, |
| "grad_norm": 29.5, |
| "learning_rate": 1.145e-05, |
| "loss": 23.1058, |
| "mean_token_accuracy": 0.6596628129482269, |
| "num_tokens": 4504001.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.4329517252743245, |
| "epoch": 0.384, |
| "grad_norm": 30.625, |
| "learning_rate": 1.195e-05, |
| "loss": 23.0101, |
| "mean_token_accuracy": 0.6596732661128044, |
| "num_tokens": 4693812.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.4356985121965409, |
| "epoch": 0.4, |
| "grad_norm": 26.0, |
| "learning_rate": 1.2450000000000003e-05, |
| "loss": 22.9403, |
| "mean_token_accuracy": 0.6609152596443891, |
| "num_tokens": 4887094.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 1.445911981910467, |
| "epoch": 0.416, |
| "grad_norm": 28.0, |
| "learning_rate": 1.295e-05, |
| "loss": 23.2032, |
| "mean_token_accuracy": 0.6575265824794769, |
| "num_tokens": 5085369.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 1.4276192136108876, |
| "epoch": 0.432, |
| "grad_norm": 30.25, |
| "learning_rate": 1.3450000000000002e-05, |
| "loss": 22.9807, |
| "mean_token_accuracy": 0.6605408634990454, |
| "num_tokens": 5271275.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 1.4254673644900322, |
| "epoch": 0.448, |
| "grad_norm": 32.0, |
| "learning_rate": 1.3950000000000002e-05, |
| "loss": 22.9563, |
| "mean_token_accuracy": 0.6602664466947317, |
| "num_tokens": 5460559.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 1.3786304764449597, |
| "epoch": 0.464, |
| "grad_norm": 28.375, |
| "learning_rate": 1.4450000000000002e-05, |
| "loss": 22.0498, |
| "mean_token_accuracy": 0.6699710045009851, |
| "num_tokens": 5653809.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 1.3980159699916839, |
| "epoch": 0.48, |
| "grad_norm": 26.0, |
| "learning_rate": 1.4950000000000003e-05, |
| "loss": 22.5067, |
| "mean_token_accuracy": 0.6643361520022154, |
| "num_tokens": 5850176.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 1.378262495994568, |
| "eval_biology_loss": 1.34170663356781, |
| "eval_biology_mean_token_accuracy": 0.6778094511032104, |
| "eval_biology_num_tokens": 5850176.0, |
| "eval_biology_runtime": 36.5741, |
| "eval_biology_samples_per_second": 13.671, |
| "eval_biology_steps_per_second": 3.418, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 1.0265604996681212, |
| "eval_chemistry_loss": 0.9883458018302917, |
| "eval_chemistry_mean_token_accuracy": 0.7520135464668274, |
| "eval_chemistry_num_tokens": 5850176.0, |
| "eval_chemistry_runtime": 44.9771, |
| "eval_chemistry_samples_per_second": 11.117, |
| "eval_chemistry_steps_per_second": 2.779, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 0.7964297127723694, |
| "eval_math_loss": 1.0260037183761597, |
| "eval_math_mean_token_accuracy": 0.759199161529541, |
| "eval_math_num_tokens": 5850176.0, |
| "eval_math_runtime": 46.2056, |
| "eval_math_samples_per_second": 10.821, |
| "eval_math_steps_per_second": 2.705, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 1.0048614959716797, |
| "eval_physics_loss": 0.9760251045227051, |
| "eval_physics_mean_token_accuracy": 0.761619930267334, |
| "eval_physics_num_tokens": 5850176.0, |
| "eval_physics_runtime": 53.0249, |
| "eval_physics_samples_per_second": 9.43, |
| "eval_physics_steps_per_second": 2.357, |
| "step": 300 |
| }, |
| { |
| "entropy": 1.3288711309432983, |
| "epoch": 0.496, |
| "grad_norm": 26.375, |
| "learning_rate": 1.545e-05, |
| "loss": 21.2235, |
| "mean_token_accuracy": 0.6774784430861474, |
| "num_tokens": 6046503.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 1.385315388813615, |
| "epoch": 0.512, |
| "grad_norm": 27.125, |
| "learning_rate": 1.595e-05, |
| "loss": 22.2247, |
| "mean_token_accuracy": 0.6680108610540628, |
| "num_tokens": 6240456.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 1.3709059737622737, |
| "epoch": 0.528, |
| "grad_norm": 24.75, |
| "learning_rate": 1.645e-05, |
| "loss": 22.1079, |
| "mean_token_accuracy": 0.6704635102301836, |
| "num_tokens": 6430555.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 1.3751415818929673, |
| "epoch": 0.544, |
| "grad_norm": 26.25, |
| "learning_rate": 1.6950000000000002e-05, |
| "loss": 21.961, |
| "mean_token_accuracy": 0.6710947025567293, |
| "num_tokens": 6626006.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 1.3602795064449311, |
| "epoch": 0.56, |
| "grad_norm": 30.625, |
| "learning_rate": 1.7450000000000004e-05, |
| "loss": 21.9474, |
| "mean_token_accuracy": 0.6700355879962444, |
| "num_tokens": 6820754.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 1.344116109609604, |
| "epoch": 0.576, |
| "grad_norm": 26.375, |
| "learning_rate": 1.795e-05, |
| "loss": 21.5572, |
| "mean_token_accuracy": 0.6739088613539934, |
| "num_tokens": 7021844.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 1.3490931887179614, |
| "epoch": 0.592, |
| "grad_norm": 26.125, |
| "learning_rate": 1.845e-05, |
| "loss": 21.7554, |
| "mean_token_accuracy": 0.6715496938675642, |
| "num_tokens": 7213951.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 1.3264936488121748, |
| "epoch": 0.608, |
| "grad_norm": 27.25, |
| "learning_rate": 1.8950000000000003e-05, |
| "loss": 21.3421, |
| "mean_token_accuracy": 0.6760978367179632, |
| "num_tokens": 7416773.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 1.348380234092474, |
| "epoch": 0.624, |
| "grad_norm": 28.125, |
| "learning_rate": 1.9450000000000002e-05, |
| "loss": 21.6293, |
| "mean_token_accuracy": 0.6746726848185063, |
| "num_tokens": 7612843.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 1.3465636018663645, |
| "epoch": 0.64, |
| "grad_norm": 26.25, |
| "learning_rate": 1.9950000000000004e-05, |
| "loss": 21.6324, |
| "mean_token_accuracy": 0.6738044265657663, |
| "num_tokens": 7801633.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 1.3016406693458558, |
| "eval_biology_loss": 1.2880380153656006, |
| "eval_biology_mean_token_accuracy": 0.6870926094055175, |
| "eval_biology_num_tokens": 7801633.0, |
| "eval_biology_runtime": 36.6216, |
| "eval_biology_samples_per_second": 13.653, |
| "eval_biology_steps_per_second": 3.413, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 0.9742186889648438, |
| "eval_chemistry_loss": 0.954098641872406, |
| "eval_chemistry_mean_token_accuracy": 0.7588451662063599, |
| "eval_chemistry_num_tokens": 7801633.0, |
| "eval_chemistry_runtime": 44.9361, |
| "eval_chemistry_samples_per_second": 11.127, |
| "eval_chemistry_steps_per_second": 2.782, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 0.7798485422134399, |
| "eval_math_loss": 1.0040466785430908, |
| "eval_math_mean_token_accuracy": 0.7620619239807129, |
| "eval_math_num_tokens": 7801633.0, |
| "eval_math_runtime": 46.2072, |
| "eval_math_samples_per_second": 10.821, |
| "eval_math_steps_per_second": 2.705, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 0.9594037117958069, |
| "eval_physics_loss": 0.9417626261711121, |
| "eval_physics_mean_token_accuracy": 0.7674873929023742, |
| "eval_physics_num_tokens": 7801633.0, |
| "eval_physics_runtime": 53.0463, |
| "eval_physics_samples_per_second": 9.426, |
| "eval_physics_steps_per_second": 2.356, |
| "step": 400 |
| }, |
| { |
| "entropy": 1.3527625199407338, |
| "epoch": 0.656, |
| "grad_norm": 27.875, |
| "learning_rate": 1.9950000000000004e-05, |
| "loss": 21.651, |
| "mean_token_accuracy": 0.6724541377276182, |
| "num_tokens": 7995843.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 1.3500655848532914, |
| "epoch": 0.672, |
| "grad_norm": 27.125, |
| "learning_rate": 1.9894444444444446e-05, |
| "loss": 21.6896, |
| "mean_token_accuracy": 0.673105026781559, |
| "num_tokens": 8183103.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 1.3231504444032907, |
| "epoch": 0.688, |
| "grad_norm": 23.5, |
| "learning_rate": 1.983888888888889e-05, |
| "loss": 21.3727, |
| "mean_token_accuracy": 0.6781029276549816, |
| "num_tokens": 8385976.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 1.3333053421229124, |
| "epoch": 0.704, |
| "grad_norm": 23.5, |
| "learning_rate": 1.9783333333333334e-05, |
| "loss": 21.2722, |
| "mean_token_accuracy": 0.6781596153974533, |
| "num_tokens": 8578431.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 1.2669135373085738, |
| "epoch": 0.72, |
| "grad_norm": 23.375, |
| "learning_rate": 1.972777777777778e-05, |
| "loss": 20.3907, |
| "mean_token_accuracy": 0.6872118715196848, |
| "num_tokens": 8781342.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 1.2877357687801123, |
| "epoch": 0.736, |
| "grad_norm": 23.875, |
| "learning_rate": 1.9672222222222222e-05, |
| "loss": 20.6999, |
| "mean_token_accuracy": 0.6825123306363821, |
| "num_tokens": 8977918.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 1.3120017491281033, |
| "epoch": 0.752, |
| "grad_norm": 24.5, |
| "learning_rate": 1.9616666666666668e-05, |
| "loss": 21.1222, |
| "mean_token_accuracy": 0.6796863108873368, |
| "num_tokens": 9169322.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 1.263747826963663, |
| "epoch": 0.768, |
| "grad_norm": 26.375, |
| "learning_rate": 1.9561111111111114e-05, |
| "loss": 20.3726, |
| "mean_token_accuracy": 0.6869186259806156, |
| "num_tokens": 9368141.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 1.3065229691565037, |
| "epoch": 0.784, |
| "grad_norm": 26.5, |
| "learning_rate": 1.9505555555555556e-05, |
| "loss": 20.8536, |
| "mean_token_accuracy": 0.683413552865386, |
| "num_tokens": 9565236.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 1.3189398009330033, |
| "epoch": 0.8, |
| "grad_norm": 25.5, |
| "learning_rate": 1.9450000000000002e-05, |
| "loss": 21.2576, |
| "mean_token_accuracy": 0.6789898939430714, |
| "num_tokens": 9761227.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 1.2551557726860045, |
| "eval_biology_loss": 1.2525163888931274, |
| "eval_biology_mean_token_accuracy": 0.6931379227638245, |
| "eval_biology_num_tokens": 9761227.0, |
| "eval_biology_runtime": 36.6065, |
| "eval_biology_samples_per_second": 13.659, |
| "eval_biology_steps_per_second": 3.415, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 0.9375595455169677, |
| "eval_chemistry_loss": 0.9276740550994873, |
| "eval_chemistry_mean_token_accuracy": 0.7639822545051574, |
| "eval_chemistry_num_tokens": 9761227.0, |
| "eval_chemistry_runtime": 44.5343, |
| "eval_chemistry_samples_per_second": 11.227, |
| "eval_chemistry_steps_per_second": 2.807, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 0.7748257055282592, |
| "eval_math_loss": 0.9847215414047241, |
| "eval_math_mean_token_accuracy": 0.7653367323875427, |
| "eval_math_num_tokens": 9761227.0, |
| "eval_math_runtime": 45.9419, |
| "eval_math_samples_per_second": 10.883, |
| "eval_math_steps_per_second": 2.721, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 0.9384980492591858, |
| "eval_physics_loss": 0.9171136021614075, |
| "eval_physics_mean_token_accuracy": 0.7720685515403748, |
| "eval_physics_num_tokens": 9761227.0, |
| "eval_physics_runtime": 52.888, |
| "eval_physics_samples_per_second": 9.454, |
| "eval_physics_steps_per_second": 2.363, |
| "step": 500 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 4000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 7, |
| "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": 1.0976936565348991e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
