| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.6, |
| "eval_steps": 100, |
| "global_step": 1000, |
| "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 |
| }, |
| { |
| "entropy": 0.9170173428952694, |
| "epoch": 0.816, |
| "grad_norm": 18.625, |
| "learning_rate": 1.8451851851851855e-05, |
| "loss": 14.5639, |
| "mean_token_accuracy": 0.7766096990555524, |
| "num_tokens": 14519893.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.9109945900738239, |
| "epoch": 0.832, |
| "grad_norm": 18.5, |
| "learning_rate": 1.8377777777777778e-05, |
| "loss": 14.6409, |
| "mean_token_accuracy": 0.7777223017066717, |
| "num_tokens": 14805088.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.9337830897420645, |
| "epoch": 0.848, |
| "grad_norm": 19.0, |
| "learning_rate": 1.8303703703703704e-05, |
| "loss": 15.0412, |
| "mean_token_accuracy": 0.7716363571584225, |
| "num_tokens": 15086306.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.9221564382314682, |
| "epoch": 0.864, |
| "grad_norm": 15.8125, |
| "learning_rate": 1.822962962962963e-05, |
| "loss": 14.7588, |
| "mean_token_accuracy": 0.7748291842639446, |
| "num_tokens": 15370985.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.9273673612624407, |
| "epoch": 0.88, |
| "grad_norm": 21.0, |
| "learning_rate": 1.8155555555555557e-05, |
| "loss": 14.9233, |
| "mean_token_accuracy": 0.7736149989068508, |
| "num_tokens": 15653836.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.9161856021732092, |
| "epoch": 0.896, |
| "grad_norm": 21.75, |
| "learning_rate": 1.8081481481481484e-05, |
| "loss": 14.7618, |
| "mean_token_accuracy": 0.7746457736939192, |
| "num_tokens": 15932179.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.8957702405750751, |
| "epoch": 0.912, |
| "grad_norm": 18.75, |
| "learning_rate": 1.800740740740741e-05, |
| "loss": 14.2902, |
| "mean_token_accuracy": 0.7794793792068958, |
| "num_tokens": 16219842.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.8888734232634306, |
| "epoch": 0.928, |
| "grad_norm": 20.875, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 14.2919, |
| "mean_token_accuracy": 0.7802723813802004, |
| "num_tokens": 16499842.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.9097447639331222, |
| "epoch": 0.944, |
| "grad_norm": 35.0, |
| "learning_rate": 1.785925925925926e-05, |
| "loss": 14.7214, |
| "mean_token_accuracy": 0.7760318968445062, |
| "num_tokens": 16781882.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.9282805230468512, |
| "epoch": 0.96, |
| "grad_norm": 16.125, |
| "learning_rate": 1.7785185185185186e-05, |
| "loss": 14.8274, |
| "mean_token_accuracy": 0.7749283302575349, |
| "num_tokens": 17067407.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 3.4800376834869384, |
| "eval_biology_loss": 3.9260165691375732, |
| "eval_biology_mean_token_accuracy": 0.41479419231414794, |
| "eval_biology_num_tokens": 17067407.0, |
| "eval_biology_runtime": 49.593, |
| "eval_biology_samples_per_second": 10.082, |
| "eval_biology_steps_per_second": 2.521, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_chemistry_entropy": 2.062796443462372, |
| "eval_chemistry_loss": 2.1318087577819824, |
| "eval_chemistry_mean_token_accuracy": 0.6000087871551514, |
| "eval_chemistry_num_tokens": 17067407.0, |
| "eval_chemistry_runtime": 61.4719, |
| "eval_chemistry_samples_per_second": 8.134, |
| "eval_chemistry_steps_per_second": 2.033, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_math_entropy": 0.9119843735694885, |
| "eval_math_loss": 0.9157347679138184, |
| "eval_math_mean_token_accuracy": 0.7754620661735535, |
| "eval_math_num_tokens": 17067407.0, |
| "eval_math_runtime": 63.0502, |
| "eval_math_samples_per_second": 7.93, |
| "eval_math_steps_per_second": 1.983, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_physics_entropy": 1.5833718237876893, |
| "eval_physics_loss": 1.6547900438308716, |
| "eval_physics_mean_token_accuracy": 0.6688490805625915, |
| "eval_physics_num_tokens": 17067407.0, |
| "eval_physics_runtime": 72.0729, |
| "eval_physics_samples_per_second": 6.937, |
| "eval_physics_steps_per_second": 1.734, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.8985319763422013, |
| "epoch": 0.976, |
| "grad_norm": 21.625, |
| "learning_rate": 1.7711111111111113e-05, |
| "loss": 14.4151, |
| "mean_token_accuracy": 0.7804601702839136, |
| "num_tokens": 17350994.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.9126239087432623, |
| "epoch": 0.992, |
| "grad_norm": 18.0, |
| "learning_rate": 1.763703703703704e-05, |
| "loss": 14.6522, |
| "mean_token_accuracy": 0.7754363786429167, |
| "num_tokens": 17637514.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.8935651954263448, |
| "epoch": 1.008, |
| "grad_norm": 17.375, |
| "learning_rate": 1.7562962962962962e-05, |
| "loss": 14.2584, |
| "mean_token_accuracy": 0.7789396952837706, |
| "num_tokens": 17926570.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.8810664782300591, |
| "epoch": 1.024, |
| "grad_norm": 17.625, |
| "learning_rate": 1.7488888888888892e-05, |
| "loss": 14.0978, |
| "mean_token_accuracy": 0.7811610117554665, |
| "num_tokens": 18207652.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.8790556967258454, |
| "epoch": 1.04, |
| "grad_norm": 20.375, |
| "learning_rate": 1.7414814814814815e-05, |
| "loss": 14.0815, |
| "mean_token_accuracy": 0.782483147084713, |
| "num_tokens": 18484931.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.872769825719297, |
| "epoch": 1.056, |
| "grad_norm": 18.625, |
| "learning_rate": 1.7340740740740742e-05, |
| "loss": 14.2329, |
| "mean_token_accuracy": 0.7814408276230097, |
| "num_tokens": 18773457.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.8854838136583567, |
| "epoch": 1.072, |
| "grad_norm": 19.125, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 14.1351, |
| "mean_token_accuracy": 0.7804106209427119, |
| "num_tokens": 19055365.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.8737944336608052, |
| "epoch": 1.088, |
| "grad_norm": 16.625, |
| "learning_rate": 1.7192592592592595e-05, |
| "loss": 14.0953, |
| "mean_token_accuracy": 0.7833730541169643, |
| "num_tokens": 19345730.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.8823883978649973, |
| "epoch": 1.104, |
| "grad_norm": 15.1875, |
| "learning_rate": 1.711851851851852e-05, |
| "loss": 14.1044, |
| "mean_token_accuracy": 0.7833204790949821, |
| "num_tokens": 19637390.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.8538309209048748, |
| "epoch": 1.12, |
| "grad_norm": 19.5, |
| "learning_rate": 1.7044444444444445e-05, |
| "loss": 13.6987, |
| "mean_token_accuracy": 0.7868878465145827, |
| "num_tokens": 19923914.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_biology_entropy": 3.443337703704834, |
| "eval_biology_loss": 3.951155662536621, |
| "eval_biology_mean_token_accuracy": 0.4163366384506226, |
| "eval_biology_num_tokens": 19923914.0, |
| "eval_biology_runtime": 49.651, |
| "eval_biology_samples_per_second": 10.07, |
| "eval_biology_steps_per_second": 2.518, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_chemistry_entropy": 2.0116386432647704, |
| "eval_chemistry_loss": 2.1324052810668945, |
| "eval_chemistry_mean_token_accuracy": 0.6028851909637452, |
| "eval_chemistry_num_tokens": 19923914.0, |
| "eval_chemistry_runtime": 61.5705, |
| "eval_chemistry_samples_per_second": 8.121, |
| "eval_chemistry_steps_per_second": 2.03, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_math_entropy": 0.8889193925857544, |
| "eval_math_loss": 0.9035109877586365, |
| "eval_math_mean_token_accuracy": 0.7773573837280273, |
| "eval_math_num_tokens": 19923914.0, |
| "eval_math_runtime": 63.1467, |
| "eval_math_samples_per_second": 7.918, |
| "eval_math_steps_per_second": 1.98, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_physics_entropy": 1.5386076803207398, |
| "eval_physics_loss": 1.6331955194473267, |
| "eval_physics_mean_token_accuracy": 0.6719050664901733, |
| "eval_physics_num_tokens": 19923914.0, |
| "eval_physics_runtime": 72.2366, |
| "eval_physics_samples_per_second": 6.922, |
| "eval_physics_steps_per_second": 1.73, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.8486799146980047, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 18.375, |
| "learning_rate": 1.697037037037037e-05, |
| "loss": 13.6728, |
| "mean_token_accuracy": 0.7864403679966927, |
| "num_tokens": 20201892.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.8742417544126511, |
| "epoch": 1.152, |
| "grad_norm": 20.0, |
| "learning_rate": 1.6896296296296298e-05, |
| "loss": 14.0206, |
| "mean_token_accuracy": 0.7843449763953686, |
| "num_tokens": 20490282.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.8681927429512143, |
| "epoch": 1.168, |
| "grad_norm": 16.625, |
| "learning_rate": 1.6822222222222224e-05, |
| "loss": 13.9888, |
| "mean_token_accuracy": 0.7844074387103319, |
| "num_tokens": 20785786.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.8594187637791038, |
| "epoch": 1.184, |
| "grad_norm": 20.875, |
| "learning_rate": 1.6748148148148147e-05, |
| "loss": 13.8341, |
| "mean_token_accuracy": 0.7861495897173881, |
| "num_tokens": 21074205.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.853638318926096, |
| "epoch": 1.2, |
| "grad_norm": 20.25, |
| "learning_rate": 1.6674074074074077e-05, |
| "loss": 13.7126, |
| "mean_token_accuracy": 0.7865742921829224, |
| "num_tokens": 21369159.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.8770630281418562, |
| "epoch": 1.216, |
| "grad_norm": 23.25, |
| "learning_rate": 1.66e-05, |
| "loss": 14.0894, |
| "mean_token_accuracy": 0.7835288234055042, |
| "num_tokens": 21649178.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.8497887661680579, |
| "epoch": 1.232, |
| "grad_norm": 19.25, |
| "learning_rate": 1.6525925925925927e-05, |
| "loss": 13.5638, |
| "mean_token_accuracy": 0.7882835403084755, |
| "num_tokens": 21930239.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.8405620498582721, |
| "epoch": 1.248, |
| "grad_norm": 17.875, |
| "learning_rate": 1.6451851851851853e-05, |
| "loss": 13.5058, |
| "mean_token_accuracy": 0.7887706879526377, |
| "num_tokens": 22216387.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.8788484372198582, |
| "epoch": 1.264, |
| "grad_norm": 20.125, |
| "learning_rate": 1.637777777777778e-05, |
| "loss": 14.209, |
| "mean_token_accuracy": 0.7804543025791645, |
| "num_tokens": 22501002.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.8771140918135643, |
| "epoch": 1.28, |
| "grad_norm": 22.25, |
| "learning_rate": 1.6303703703703706e-05, |
| "loss": 13.9869, |
| "mean_token_accuracy": 0.7842618364840745, |
| "num_tokens": 22779682.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_biology_entropy": 3.436663423538208, |
| "eval_biology_loss": 3.9039478302001953, |
| "eval_biology_mean_token_accuracy": 0.41795943784713746, |
| "eval_biology_num_tokens": 22779682.0, |
| "eval_biology_runtime": 49.6843, |
| "eval_biology_samples_per_second": 10.064, |
| "eval_biology_steps_per_second": 2.516, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_chemistry_entropy": 1.97736017036438, |
| "eval_chemistry_loss": 2.080076217651367, |
| "eval_chemistry_mean_token_accuracy": 0.606924372434616, |
| "eval_chemistry_num_tokens": 22779682.0, |
| "eval_chemistry_runtime": 61.5504, |
| "eval_chemistry_samples_per_second": 8.123, |
| "eval_chemistry_steps_per_second": 2.031, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_math_entropy": 0.8625033555030823, |
| "eval_math_loss": 0.892142653465271, |
| "eval_math_mean_token_accuracy": 0.7796395468711853, |
| "eval_math_num_tokens": 22779682.0, |
| "eval_math_runtime": 63.1583, |
| "eval_math_samples_per_second": 7.917, |
| "eval_math_steps_per_second": 1.979, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_physics_entropy": 1.5225397615432739, |
| "eval_physics_loss": 1.612709879875183, |
| "eval_physics_mean_token_accuracy": 0.674011738061905, |
| "eval_physics_num_tokens": 22779682.0, |
| "eval_physics_runtime": 72.1546, |
| "eval_physics_samples_per_second": 6.93, |
| "eval_physics_steps_per_second": 1.732, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.8474511569365859, |
| "epoch": 1.296, |
| "grad_norm": 17.75, |
| "learning_rate": 1.622962962962963e-05, |
| "loss": 13.6822, |
| "mean_token_accuracy": 0.7866408120840788, |
| "num_tokens": 23057744.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.8580924984067678, |
| "epoch": 1.312, |
| "grad_norm": 21.5, |
| "learning_rate": 1.6155555555555556e-05, |
| "loss": 13.9676, |
| "mean_token_accuracy": 0.7842281598597765, |
| "num_tokens": 23344644.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.8850870780646801, |
| "epoch": 1.328, |
| "grad_norm": 21.5, |
| "learning_rate": 1.6081481481481482e-05, |
| "loss": 14.0615, |
| "mean_token_accuracy": 0.781786335259676, |
| "num_tokens": 23622405.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.8474001951515675, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 18.25, |
| "learning_rate": 1.600740740740741e-05, |
| "loss": 13.6058, |
| "mean_token_accuracy": 0.7873330093920231, |
| "num_tokens": 23899771.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.8435361562296748, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 16.0, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 13.6319, |
| "mean_token_accuracy": 0.7882646590471267, |
| "num_tokens": 24187023.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.9156765401363373, |
| "epoch": 1.376, |
| "grad_norm": 21.5, |
| "learning_rate": 1.5859259259259262e-05, |
| "loss": 14.6843, |
| "mean_token_accuracy": 0.7743376530706882, |
| "num_tokens": 24466132.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.8523435527458787, |
| "epoch": 1.392, |
| "grad_norm": 20.75, |
| "learning_rate": 1.5785185185185185e-05, |
| "loss": 13.6998, |
| "mean_token_accuracy": 0.7863496944308281, |
| "num_tokens": 24748043.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.8375731509178876, |
| "epoch": 1.408, |
| "grad_norm": 21.125, |
| "learning_rate": 1.571111111111111e-05, |
| "loss": 13.5188, |
| "mean_token_accuracy": 0.7898116644471884, |
| "num_tokens": 25036674.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.8376417780295015, |
| "epoch": 1.424, |
| "grad_norm": 15.25, |
| "learning_rate": 1.5637037037037038e-05, |
| "loss": 13.4567, |
| "mean_token_accuracy": 0.7908716473728419, |
| "num_tokens": 25324579.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.8603908697143197, |
| "epoch": 1.44, |
| "grad_norm": 19.125, |
| "learning_rate": 1.5562962962962965e-05, |
| "loss": 13.9612, |
| "mean_token_accuracy": 0.7841609816998243, |
| "num_tokens": 25606824.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_biology_entropy": 3.45110001373291, |
| "eval_biology_loss": 3.8754560947418213, |
| "eval_biology_mean_token_accuracy": 0.4207558875083923, |
| "eval_biology_num_tokens": 25606824.0, |
| "eval_biology_runtime": 49.6588, |
| "eval_biology_samples_per_second": 10.069, |
| "eval_biology_steps_per_second": 2.517, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_chemistry_entropy": 1.9984806547164917, |
| "eval_chemistry_loss": 2.088613748550415, |
| "eval_chemistry_mean_token_accuracy": 0.607667599439621, |
| "eval_chemistry_num_tokens": 25606824.0, |
| "eval_chemistry_runtime": 61.604, |
| "eval_chemistry_samples_per_second": 8.116, |
| "eval_chemistry_steps_per_second": 2.029, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_math_entropy": 0.8740571973323822, |
| "eval_math_loss": 0.8836535811424255, |
| "eval_math_mean_token_accuracy": 0.7810649185180664, |
| "eval_math_num_tokens": 25606824.0, |
| "eval_math_runtime": 63.1568, |
| "eval_math_samples_per_second": 7.917, |
| "eval_math_steps_per_second": 1.979, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_physics_entropy": 1.533125587463379, |
| "eval_physics_loss": 1.6004778146743774, |
| "eval_physics_mean_token_accuracy": 0.6762587761878968, |
| "eval_physics_num_tokens": 25606824.0, |
| "eval_physics_runtime": 71.9331, |
| "eval_physics_samples_per_second": 6.951, |
| "eval_physics_steps_per_second": 1.738, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.8511543165892362, |
| "epoch": 1.456, |
| "grad_norm": 16.125, |
| "learning_rate": 1.548888888888889e-05, |
| "loss": 13.5133, |
| "mean_token_accuracy": 0.7876119159162045, |
| "num_tokens": 25886396.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.8463271845132112, |
| "epoch": 1.472, |
| "grad_norm": 17.75, |
| "learning_rate": 1.5414814814814814e-05, |
| "loss": 13.5862, |
| "mean_token_accuracy": 0.787867683917284, |
| "num_tokens": 26163618.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.8713837819173932, |
| "epoch": 1.488, |
| "grad_norm": 20.125, |
| "learning_rate": 1.5340740740740744e-05, |
| "loss": 14.1889, |
| "mean_token_accuracy": 0.7813853111118079, |
| "num_tokens": 26438338.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.834681292809546, |
| "epoch": 1.504, |
| "grad_norm": 21.875, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 13.4447, |
| "mean_token_accuracy": 0.7905492588877678, |
| "num_tokens": 26729255.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.8771929260343313, |
| "epoch": 1.52, |
| "grad_norm": 18.0, |
| "learning_rate": 1.5192592592592594e-05, |
| "loss": 14.0064, |
| "mean_token_accuracy": 0.7831294402480126, |
| "num_tokens": 27017935.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.8460531679913401, |
| "epoch": 1.536, |
| "grad_norm": 16.125, |
| "learning_rate": 1.5118518518518519e-05, |
| "loss": 13.6594, |
| "mean_token_accuracy": 0.786826192215085, |
| "num_tokens": 27306339.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.8540312614291906, |
| "epoch": 1.552, |
| "grad_norm": 18.75, |
| "learning_rate": 1.5044444444444445e-05, |
| "loss": 13.6638, |
| "mean_token_accuracy": 0.7870731830596924, |
| "num_tokens": 27591959.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.8264467475935817, |
| "epoch": 1.568, |
| "grad_norm": 16.125, |
| "learning_rate": 1.497037037037037e-05, |
| "loss": 13.3238, |
| "mean_token_accuracy": 0.7920118164271116, |
| "num_tokens": 27884398.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.8386921418830753, |
| "epoch": 1.584, |
| "grad_norm": 20.25, |
| "learning_rate": 1.4896296296296298e-05, |
| "loss": 13.5938, |
| "mean_token_accuracy": 0.7884967364370823, |
| "num_tokens": 28171274.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.8454421229660511, |
| "epoch": 1.6, |
| "grad_norm": 21.125, |
| "learning_rate": 1.4822222222222225e-05, |
| "loss": 13.4115, |
| "mean_token_accuracy": 0.7884562328457833, |
| "num_tokens": 28457624.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_biology_entropy": 3.4279630908966063, |
| "eval_biology_loss": 3.9252524375915527, |
| "eval_biology_mean_token_accuracy": 0.42100021028518675, |
| "eval_biology_num_tokens": 28457624.0, |
| "eval_biology_runtime": 49.5484, |
| "eval_biology_samples_per_second": 10.091, |
| "eval_biology_steps_per_second": 2.523, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_chemistry_entropy": 1.9675587797164917, |
| "eval_chemistry_loss": 2.0829849243164062, |
| "eval_chemistry_mean_token_accuracy": 0.6087198803424835, |
| "eval_chemistry_num_tokens": 28457624.0, |
| "eval_chemistry_runtime": 60.8881, |
| "eval_chemistry_samples_per_second": 8.212, |
| "eval_chemistry_steps_per_second": 2.053, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_math_entropy": 0.858300416469574, |
| "eval_math_loss": 0.8771592974662781, |
| "eval_math_mean_token_accuracy": 0.7824492840766907, |
| "eval_math_num_tokens": 28457624.0, |
| "eval_math_runtime": 62.7471, |
| "eval_math_samples_per_second": 7.968, |
| "eval_math_steps_per_second": 1.992, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_physics_entropy": 1.5079545302391053, |
| "eval_physics_loss": 1.5952943563461304, |
| "eval_physics_mean_token_accuracy": 0.6772087650299072, |
| "eval_physics_num_tokens": 28457624.0, |
| "eval_physics_runtime": 71.6735, |
| "eval_physics_samples_per_second": 6.976, |
| "eval_physics_steps_per_second": 1.744, |
| "step": 1000 |
| } |
| ], |
| "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": 2.897870312538204e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
