| { |
| "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": 1.1280501924455166, |
| "epoch": 0.016, |
| "grad_norm": 161.0, |
| "learning_rate": 6.000000000000001e-07, |
| "loss": 36.0968, |
| "mean_token_accuracy": 0.6165979273617268, |
| "num_tokens": 280941.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.1705728020519017, |
| "epoch": 0.032, |
| "grad_norm": 146.0, |
| "learning_rate": 1.2666666666666669e-06, |
| "loss": 36.5572, |
| "mean_token_accuracy": 0.6121777599677444, |
| "num_tokens": 558056.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.1892314109951259, |
| "epoch": 0.048, |
| "grad_norm": 135.0, |
| "learning_rate": 1.9333333333333336e-06, |
| "loss": 35.5667, |
| "mean_token_accuracy": 0.6162483215332031, |
| "num_tokens": 836753.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.2923907738178968, |
| "epoch": 0.064, |
| "grad_norm": 106.0, |
| "learning_rate": 2.6e-06, |
| "loss": 34.4747, |
| "mean_token_accuracy": 0.6145394820719957, |
| "num_tokens": 1126446.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.4465828079730272, |
| "epoch": 0.08, |
| "grad_norm": 86.5, |
| "learning_rate": 3.266666666666667e-06, |
| "loss": 32.2616, |
| "mean_token_accuracy": 0.6272102266550064, |
| "num_tokens": 1413596.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.5344239331781864, |
| "epoch": 0.096, |
| "grad_norm": 54.25, |
| "learning_rate": 3.9333333333333335e-06, |
| "loss": 28.8152, |
| "mean_token_accuracy": 0.6551170501857996, |
| "num_tokens": 1701193.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.5569457441568375, |
| "epoch": 0.112, |
| "grad_norm": 45.75, |
| "learning_rate": 4.600000000000001e-06, |
| "loss": 26.6094, |
| "mean_token_accuracy": 0.6667167004197836, |
| "num_tokens": 1979232.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.4980619736015797, |
| "epoch": 0.128, |
| "grad_norm": 34.0, |
| "learning_rate": 5.2666666666666665e-06, |
| "loss": 25.0735, |
| "mean_token_accuracy": 0.6823939502239227, |
| "num_tokens": 2274177.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.4427384316921235, |
| "epoch": 0.144, |
| "grad_norm": 29.75, |
| "learning_rate": 5.933333333333335e-06, |
| "loss": 23.2103, |
| "mean_token_accuracy": 0.6962548345327377, |
| "num_tokens": 2548445.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.3565625827759504, |
| "epoch": 0.16, |
| "grad_norm": 30.875, |
| "learning_rate": 6.600000000000001e-06, |
| "loss": 21.4845, |
| "mean_token_accuracy": 0.7108539894223214, |
| "num_tokens": 2824418.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 3.535919818878174, |
| "eval_biology_loss": 5.02457332611084, |
| "eval_biology_mean_token_accuracy": 0.3542800180912018, |
| "eval_biology_num_tokens": 2824418.0, |
| "eval_biology_runtime": 49.6437, |
| "eval_biology_samples_per_second": 10.072, |
| "eval_biology_steps_per_second": 2.518, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 2.5280555896759034, |
| "eval_chemistry_loss": 2.8887903690338135, |
| "eval_chemistry_mean_token_accuracy": 0.5300401592254639, |
| "eval_chemistry_num_tokens": 2824418.0, |
| "eval_chemistry_runtime": 61.5471, |
| "eval_chemistry_samples_per_second": 8.124, |
| "eval_chemistry_steps_per_second": 2.031, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 1.3736235461235047, |
| "eval_math_loss": 1.3470827341079712, |
| "eval_math_mean_token_accuracy": 0.7081305732727051, |
| "eval_math_num_tokens": 2824418.0, |
| "eval_math_runtime": 63.1282, |
| "eval_math_samples_per_second": 7.92, |
| "eval_math_steps_per_second": 1.98, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 2.064271595954895, |
| "eval_physics_loss": 2.1881864070892334, |
| "eval_physics_mean_token_accuracy": 0.6055651969909668, |
| "eval_physics_num_tokens": 2824418.0, |
| "eval_physics_runtime": 71.8769, |
| "eval_physics_samples_per_second": 6.956, |
| "eval_physics_steps_per_second": 1.739, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.3322576396167278, |
| "epoch": 0.176, |
| "grad_norm": 23.25, |
| "learning_rate": 7.266666666666668e-06, |
| "loss": 20.9108, |
| "mean_token_accuracy": 0.7151085637509823, |
| "num_tokens": 3110313.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.3247066088020802, |
| "epoch": 0.192, |
| "grad_norm": 23.625, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 20.6356, |
| "mean_token_accuracy": 0.7142969533801079, |
| "num_tokens": 3394170.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.256456857919693, |
| "epoch": 0.208, |
| "grad_norm": 21.125, |
| "learning_rate": 8.6e-06, |
| "loss": 19.9583, |
| "mean_token_accuracy": 0.7226813279092312, |
| "num_tokens": 3673600.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.2233942866325378, |
| "epoch": 0.224, |
| "grad_norm": 23.5, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 19.4829, |
| "mean_token_accuracy": 0.7258789885789156, |
| "num_tokens": 3953732.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.1788023415952922, |
| "epoch": 0.24, |
| "grad_norm": 22.625, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 18.6617, |
| "mean_token_accuracy": 0.735758152604103, |
| "num_tokens": 4243655.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.1358939699828625, |
| "epoch": 0.256, |
| "grad_norm": 21.25, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 18.067, |
| "mean_token_accuracy": 0.7414251279085875, |
| "num_tokens": 4531471.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.1413200695067645, |
| "epoch": 0.272, |
| "grad_norm": 18.75, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 18.2054, |
| "mean_token_accuracy": 0.740831508487463, |
| "num_tokens": 4810284.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.125382575765252, |
| "epoch": 0.288, |
| "grad_norm": 17.875, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 17.7696, |
| "mean_token_accuracy": 0.7443546559661627, |
| "num_tokens": 5095104.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.1154642656445504, |
| "epoch": 0.304, |
| "grad_norm": 17.125, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 17.7511, |
| "mean_token_accuracy": 0.7442465547472239, |
| "num_tokens": 5383732.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.0828186847269534, |
| "epoch": 0.32, |
| "grad_norm": 19.875, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 17.2671, |
| "mean_token_accuracy": 0.7503975994884968, |
| "num_tokens": 5676334.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 3.6005834312438965, |
| "eval_biology_loss": 4.470179557800293, |
| "eval_biology_mean_token_accuracy": 0.3823731031417847, |
| "eval_biology_num_tokens": 5676334.0, |
| "eval_biology_runtime": 49.5338, |
| "eval_biology_samples_per_second": 10.094, |
| "eval_biology_steps_per_second": 2.524, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 2.351666664123535, |
| "eval_chemistry_loss": 2.5015459060668945, |
| "eval_chemistry_mean_token_accuracy": 0.5612157773971558, |
| "eval_chemistry_num_tokens": 5676334.0, |
| "eval_chemistry_runtime": 61.4417, |
| "eval_chemistry_samples_per_second": 8.138, |
| "eval_chemistry_steps_per_second": 2.034, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 1.1107734112739562, |
| "eval_math_loss": 1.1035394668579102, |
| "eval_math_mean_token_accuracy": 0.744808388710022, |
| "eval_math_num_tokens": 5676334.0, |
| "eval_math_runtime": 63.0675, |
| "eval_math_samples_per_second": 7.928, |
| "eval_math_steps_per_second": 1.982, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 1.8196128549575805, |
| "eval_physics_loss": 1.8889656066894531, |
| "eval_physics_mean_token_accuracy": 0.6360980491638184, |
| "eval_physics_num_tokens": 5676334.0, |
| "eval_physics_runtime": 71.7937, |
| "eval_physics_samples_per_second": 6.964, |
| "eval_physics_steps_per_second": 1.741, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.0973670542240144, |
| "epoch": 0.336, |
| "grad_norm": 19.0, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 17.6111, |
| "mean_token_accuracy": 0.7442069798707962, |
| "num_tokens": 5958480.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.064722668007016, |
| "epoch": 0.352, |
| "grad_norm": 18.25, |
| "learning_rate": 1.46e-05, |
| "loss": 16.8683, |
| "mean_token_accuracy": 0.752158772200346, |
| "num_tokens": 6242161.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 1.0617528311908244, |
| "epoch": 0.368, |
| "grad_norm": 20.5, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 16.9997, |
| "mean_token_accuracy": 0.751146937161684, |
| "num_tokens": 6523679.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.0443669583648443, |
| "epoch": 0.384, |
| "grad_norm": 19.625, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 16.6591, |
| "mean_token_accuracy": 0.7551476296037436, |
| "num_tokens": 6810978.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.0753562293946743, |
| "epoch": 0.4, |
| "grad_norm": 17.875, |
| "learning_rate": 1.66e-05, |
| "loss": 17.4353, |
| "mean_token_accuracy": 0.7485508523881436, |
| "num_tokens": 7096903.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 1.0416356481611728, |
| "epoch": 0.416, |
| "grad_norm": 17.375, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 16.623, |
| "mean_token_accuracy": 0.7539484884589911, |
| "num_tokens": 7377181.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 1.0091398146003485, |
| "epoch": 0.432, |
| "grad_norm": 20.375, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 16.1553, |
| "mean_token_accuracy": 0.7619809154421091, |
| "num_tokens": 7650523.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.9927664395421744, |
| "epoch": 0.448, |
| "grad_norm": 20.75, |
| "learning_rate": 1.86e-05, |
| "loss": 15.9549, |
| "mean_token_accuracy": 0.7623589888215065, |
| "num_tokens": 7936788.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.9981700140982867, |
| "epoch": 0.464, |
| "grad_norm": 21.75, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 16.1264, |
| "mean_token_accuracy": 0.761608887091279, |
| "num_tokens": 8223147.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 1.0235895983874799, |
| "epoch": 0.48, |
| "grad_norm": 19.375, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 16.2896, |
| "mean_token_accuracy": 0.7603169117122889, |
| "num_tokens": 8506030.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 3.66986682510376, |
| "eval_biology_loss": 4.148224353790283, |
| "eval_biology_mean_token_accuracy": 0.3987880692481995, |
| "eval_biology_num_tokens": 8506030.0, |
| "eval_biology_runtime": 49.5301, |
| "eval_biology_samples_per_second": 10.095, |
| "eval_biology_steps_per_second": 2.524, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 2.309468500137329, |
| "eval_chemistry_loss": 2.3434462547302246, |
| "eval_chemistry_mean_token_accuracy": 0.5782731935977936, |
| "eval_chemistry_num_tokens": 8506030.0, |
| "eval_chemistry_runtime": 61.4332, |
| "eval_chemistry_samples_per_second": 8.139, |
| "eval_chemistry_steps_per_second": 2.035, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 1.007653223514557, |
| "eval_math_loss": 1.0127896070480347, |
| "eval_math_mean_token_accuracy": 0.7593745794296265, |
| "eval_math_num_tokens": 8506030.0, |
| "eval_math_runtime": 63.0217, |
| "eval_math_samples_per_second": 7.934, |
| "eval_math_steps_per_second": 1.983, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 1.7608964786529542, |
| "eval_physics_loss": 1.7697628736495972, |
| "eval_physics_mean_token_accuracy": 0.6516191956996917, |
| "eval_physics_num_tokens": 8506030.0, |
| "eval_physics_runtime": 71.8045, |
| "eval_physics_samples_per_second": 6.963, |
| "eval_physics_steps_per_second": 1.741, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.9751624643802643, |
| "epoch": 0.496, |
| "grad_norm": 27.0, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 15.6472, |
| "mean_token_accuracy": 0.7660828057676554, |
| "num_tokens": 8788726.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.9841579314321279, |
| "epoch": 0.512, |
| "grad_norm": 19.5, |
| "learning_rate": 1.985925925925926e-05, |
| "loss": 15.7341, |
| "mean_token_accuracy": 0.765475382655859, |
| "num_tokens": 9078039.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.9567954104393721, |
| "epoch": 0.528, |
| "grad_norm": 22.875, |
| "learning_rate": 1.9785185185185187e-05, |
| "loss": 15.3963, |
| "mean_token_accuracy": 0.7705613534897566, |
| "num_tokens": 9373860.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.9622254882007837, |
| "epoch": 0.544, |
| "grad_norm": 20.625, |
| "learning_rate": 1.971111111111111e-05, |
| "loss": 15.6284, |
| "mean_token_accuracy": 0.7682077161967754, |
| "num_tokens": 9660940.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 1.013891428336501, |
| "epoch": 0.56, |
| "grad_norm": 28.875, |
| "learning_rate": 1.963703703703704e-05, |
| "loss": 16.1123, |
| "mean_token_accuracy": 0.758896803110838, |
| "num_tokens": 9932302.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.9643709316849709, |
| "epoch": 0.576, |
| "grad_norm": 20.0, |
| "learning_rate": 1.9562962962962964e-05, |
| "loss": 15.4925, |
| "mean_token_accuracy": 0.7669644717127084, |
| "num_tokens": 10215462.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.9496656563133001, |
| "epoch": 0.592, |
| "grad_norm": 18.5, |
| "learning_rate": 1.948888888888889e-05, |
| "loss": 15.3772, |
| "mean_token_accuracy": 0.7689070332795381, |
| "num_tokens": 10504396.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.9635139308869839, |
| "epoch": 0.608, |
| "grad_norm": 17.5, |
| "learning_rate": 1.9414814814814817e-05, |
| "loss": 15.4665, |
| "mean_token_accuracy": 0.768926914036274, |
| "num_tokens": 10793126.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.9394328493624926, |
| "epoch": 0.624, |
| "grad_norm": 19.375, |
| "learning_rate": 1.9340740740740743e-05, |
| "loss": 15.1417, |
| "mean_token_accuracy": 0.7703750677406788, |
| "num_tokens": 11081768.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.930184380710125, |
| "epoch": 0.64, |
| "grad_norm": 18.25, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 14.8657, |
| "mean_token_accuracy": 0.7726314965635538, |
| "num_tokens": 11370320.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 3.578768089294434, |
| "eval_biology_loss": 4.099727630615234, |
| "eval_biology_mean_token_accuracy": 0.4042117977142334, |
| "eval_biology_num_tokens": 11370320.0, |
| "eval_biology_runtime": 49.5083, |
| "eval_biology_samples_per_second": 10.099, |
| "eval_biology_steps_per_second": 2.525, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 2.1963028135299685, |
| "eval_chemistry_loss": 2.250903367996216, |
| "eval_chemistry_mean_token_accuracy": 0.588011109828949, |
| "eval_chemistry_num_tokens": 11370320.0, |
| "eval_chemistry_runtime": 61.4213, |
| "eval_chemistry_samples_per_second": 8.14, |
| "eval_chemistry_steps_per_second": 2.035, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 0.9574988222122193, |
| "eval_math_loss": 0.9642173051834106, |
| "eval_math_mean_token_accuracy": 0.7676870808601379, |
| "eval_math_num_tokens": 11370320.0, |
| "eval_math_runtime": 62.9918, |
| "eval_math_samples_per_second": 7.938, |
| "eval_math_steps_per_second": 1.984, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 1.6637140741348266, |
| "eval_physics_loss": 1.7104088068008423, |
| "eval_physics_mean_token_accuracy": 0.6602483534812927, |
| "eval_physics_num_tokens": 11370320.0, |
| "eval_physics_runtime": 71.7644, |
| "eval_physics_samples_per_second": 6.967, |
| "eval_physics_steps_per_second": 1.742, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.908730499073863, |
| "epoch": 0.656, |
| "grad_norm": 16.625, |
| "learning_rate": 1.9192592592592593e-05, |
| "loss": 14.6477, |
| "mean_token_accuracy": 0.7755995105952025, |
| "num_tokens": 11657835.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.9500968560576439, |
| "epoch": 0.672, |
| "grad_norm": 18.25, |
| "learning_rate": 1.911851851851852e-05, |
| "loss": 15.2383, |
| "mean_token_accuracy": 0.769153194501996, |
| "num_tokens": 11949262.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.9080810055136681, |
| "epoch": 0.688, |
| "grad_norm": 15.625, |
| "learning_rate": 1.9044444444444446e-05, |
| "loss": 14.6139, |
| "mean_token_accuracy": 0.7764004003256559, |
| "num_tokens": 12227640.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.9170140367001295, |
| "epoch": 0.704, |
| "grad_norm": 17.625, |
| "learning_rate": 1.8970370370370372e-05, |
| "loss": 14.7655, |
| "mean_token_accuracy": 0.7752966023981571, |
| "num_tokens": 12516641.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.9368526700884103, |
| "epoch": 0.72, |
| "grad_norm": 19.25, |
| "learning_rate": 1.8896296296296295e-05, |
| "loss": 15.134, |
| "mean_token_accuracy": 0.7706776697188615, |
| "num_tokens": 12793343.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.9261821806430817, |
| "epoch": 0.736, |
| "grad_norm": 17.5, |
| "learning_rate": 1.8822222222222225e-05, |
| "loss": 14.833, |
| "mean_token_accuracy": 0.7745088890194893, |
| "num_tokens": 13077981.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.9040410481393337, |
| "epoch": 0.752, |
| "grad_norm": 19.75, |
| "learning_rate": 1.874814814814815e-05, |
| "loss": 14.4907, |
| "mean_token_accuracy": 0.7791126936674118, |
| "num_tokens": 13358957.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.8739457067102194, |
| "epoch": 0.768, |
| "grad_norm": 15.8125, |
| "learning_rate": 1.8674074074074075e-05, |
| "loss": 14.112, |
| "mean_token_accuracy": 0.7853138707578182, |
| "num_tokens": 13653412.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.946398203074932, |
| "epoch": 0.784, |
| "grad_norm": 19.875, |
| "learning_rate": 1.86e-05, |
| "loss": 15.145, |
| "mean_token_accuracy": 0.7723897945135831, |
| "num_tokens": 13940856.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.9361972361803055, |
| "epoch": 0.8, |
| "grad_norm": 19.625, |
| "learning_rate": 1.8525925925925928e-05, |
| "loss": 15.2105, |
| "mean_token_accuracy": 0.7712519209831953, |
| "num_tokens": 14230754.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 3.714933115005493, |
| "eval_biology_loss": 3.970184564590454, |
| "eval_biology_mean_token_accuracy": 0.41206827068328855, |
| "eval_biology_num_tokens": 14230754.0, |
| "eval_biology_runtime": 49.5281, |
| "eval_biology_samples_per_second": 10.095, |
| "eval_biology_steps_per_second": 2.524, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 2.237494807243347, |
| "eval_chemistry_loss": 2.1822235584259033, |
| "eval_chemistry_mean_token_accuracy": 0.5956678519248962, |
| "eval_chemistry_num_tokens": 14230754.0, |
| "eval_chemistry_runtime": 61.0757, |
| "eval_chemistry_samples_per_second": 8.187, |
| "eval_chemistry_steps_per_second": 2.047, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 0.9503358464241027, |
| "eval_math_loss": 0.9351838827133179, |
| "eval_math_mean_token_accuracy": 0.7728546581268311, |
| "eval_math_num_tokens": 14230754.0, |
| "eval_math_runtime": 63.0753, |
| "eval_math_samples_per_second": 7.927, |
| "eval_math_steps_per_second": 1.982, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 1.6920014672279358, |
| "eval_physics_loss": 1.671562671661377, |
| "eval_physics_mean_token_accuracy": 0.6658579094409942, |
| "eval_physics_num_tokens": 14230754.0, |
| "eval_physics_runtime": 71.8921, |
| "eval_physics_samples_per_second": 6.955, |
| "eval_physics_steps_per_second": 1.739, |
| "step": 500 |
| } |
| ], |
| "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": 1.448451560178856e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
