| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 4.0, |
| "eval_steps": 100, |
| "global_step": 2500, |
| "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 |
| }, |
| { |
| "entropy": 0.8404561445116997, |
| "epoch": 1.616, |
| "grad_norm": 17.0, |
| "learning_rate": 1.474814814814815e-05, |
| "loss": 13.6717, |
| "mean_token_accuracy": 0.7868105556815863, |
| "num_tokens": 28743099.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.8566132893785834, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 19.0, |
| "learning_rate": 1.4674074074074076e-05, |
| "loss": 13.6689, |
| "mean_token_accuracy": 0.7859214432537556, |
| "num_tokens": 29017297.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.8290193160995841, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 19.5, |
| "learning_rate": 1.46e-05, |
| "loss": 13.341, |
| "mean_token_accuracy": 0.7918279275298119, |
| "num_tokens": 29303707.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.8359301598742604, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 19.25, |
| "learning_rate": 1.4525925925925927e-05, |
| "loss": 13.4233, |
| "mean_token_accuracy": 0.7886226952075959, |
| "num_tokens": 29597156.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.8554202366620303, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 17.875, |
| "learning_rate": 1.4451851851851852e-05, |
| "loss": 13.7053, |
| "mean_token_accuracy": 0.7881614901125431, |
| "num_tokens": 29883879.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.8544707661494613, |
| "epoch": 1.696, |
| "grad_norm": 19.25, |
| "learning_rate": 1.4377777777777779e-05, |
| "loss": 13.8386, |
| "mean_token_accuracy": 0.7859702635556459, |
| "num_tokens": 30165760.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.8261727832257748, |
| "epoch": 1.712, |
| "grad_norm": 17.875, |
| "learning_rate": 1.4303703703703703e-05, |
| "loss": 13.3729, |
| "mean_token_accuracy": 0.7911275729537011, |
| "num_tokens": 30460367.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.8331544175744057, |
| "epoch": 1.728, |
| "grad_norm": 24.875, |
| "learning_rate": 1.4229629629629632e-05, |
| "loss": 13.2888, |
| "mean_token_accuracy": 0.7924999501556158, |
| "num_tokens": 30739137.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.8388408731669188, |
| "epoch": 1.744, |
| "grad_norm": 19.375, |
| "learning_rate": 1.4155555555555556e-05, |
| "loss": 13.4587, |
| "mean_token_accuracy": 0.7895076658576727, |
| "num_tokens": 31022663.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.8255633022636175, |
| "epoch": 1.76, |
| "grad_norm": 18.75, |
| "learning_rate": 1.4081481481481483e-05, |
| "loss": 13.2975, |
| "mean_token_accuracy": 0.7897602036595345, |
| "num_tokens": 31306494.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_biology_entropy": 3.4277304420471193, |
| "eval_biology_loss": 3.8936383724212646, |
| "eval_biology_mean_token_accuracy": 0.421789103269577, |
| "eval_biology_num_tokens": 31306494.0, |
| "eval_biology_runtime": 49.7293, |
| "eval_biology_samples_per_second": 10.054, |
| "eval_biology_steps_per_second": 2.514, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_chemistry_entropy": 1.9663724312782287, |
| "eval_chemistry_loss": 2.0707342624664307, |
| "eval_chemistry_mean_token_accuracy": 0.6100191533565521, |
| "eval_chemistry_num_tokens": 31306494.0, |
| "eval_chemistry_runtime": 61.6224, |
| "eval_chemistry_samples_per_second": 8.114, |
| "eval_chemistry_steps_per_second": 2.028, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_math_entropy": 0.8585206677913666, |
| "eval_math_loss": 0.8713729381561279, |
| "eval_math_mean_token_accuracy": 0.7834768991470337, |
| "eval_math_num_tokens": 31306494.0, |
| "eval_math_runtime": 63.1881, |
| "eval_math_samples_per_second": 7.913, |
| "eval_math_steps_per_second": 1.978, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_physics_entropy": 1.5165297889709473, |
| "eval_physics_loss": 1.5930066108703613, |
| "eval_physics_mean_token_accuracy": 0.6775295588970184, |
| "eval_physics_num_tokens": 31306494.0, |
| "eval_physics_runtime": 71.9549, |
| "eval_physics_samples_per_second": 6.949, |
| "eval_physics_steps_per_second": 1.737, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.8478734942153097, |
| "epoch": 1.776, |
| "grad_norm": 19.375, |
| "learning_rate": 1.400740740740741e-05, |
| "loss": 13.6126, |
| "mean_token_accuracy": 0.7887870900332927, |
| "num_tokens": 31595542.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.8505856825038791, |
| "epoch": 1.792, |
| "grad_norm": 18.0, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 13.7357, |
| "mean_token_accuracy": 0.785324577614665, |
| "num_tokens": 31881189.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.8546114789322019, |
| "epoch": 1.808, |
| "grad_norm": 17.625, |
| "learning_rate": 1.385925925925926e-05, |
| "loss": 13.6904, |
| "mean_token_accuracy": 0.7876242090016603, |
| "num_tokens": 32164196.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.7973219279199839, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 18.375, |
| "learning_rate": 1.3785185185185186e-05, |
| "loss": 12.8257, |
| "mean_token_accuracy": 0.7969068605452776, |
| "num_tokens": 32441530.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.8270328000187874, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 20.0, |
| "learning_rate": 1.3711111111111112e-05, |
| "loss": 13.348, |
| "mean_token_accuracy": 0.791271859779954, |
| "num_tokens": 32723145.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.8601775204762816, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 17.125, |
| "learning_rate": 1.3637037037037037e-05, |
| "loss": 13.7437, |
| "mean_token_accuracy": 0.7868250191211701, |
| "num_tokens": 33011263.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.8270715646445751, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 17.375, |
| "learning_rate": 1.3562962962962965e-05, |
| "loss": 13.3305, |
| "mean_token_accuracy": 0.7928860258311033, |
| "num_tokens": 33298921.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.8495924055576325, |
| "epoch": 1.888, |
| "grad_norm": 20.375, |
| "learning_rate": 1.3488888888888888e-05, |
| "loss": 13.5645, |
| "mean_token_accuracy": 0.7878689642995595, |
| "num_tokens": 33576243.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.8353265527635813, |
| "epoch": 1.904, |
| "grad_norm": 17.0, |
| "learning_rate": 1.3414814814814817e-05, |
| "loss": 13.49, |
| "mean_token_accuracy": 0.7885459426790475, |
| "num_tokens": 33850968.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.8119674637913704, |
| "epoch": 1.92, |
| "grad_norm": 17.0, |
| "learning_rate": 1.3340740740740741e-05, |
| "loss": 13.0727, |
| "mean_token_accuracy": 0.7935783289372921, |
| "num_tokens": 34128558.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_biology_entropy": 3.3561630630493164, |
| "eval_biology_loss": 3.8274106979370117, |
| "eval_biology_mean_token_accuracy": 0.42412702322006224, |
| "eval_biology_num_tokens": 34128558.0, |
| "eval_biology_runtime": 50.4341, |
| "eval_biology_samples_per_second": 9.914, |
| "eval_biology_steps_per_second": 2.478, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_chemistry_entropy": 1.9330943393707276, |
| "eval_chemistry_loss": 2.0680081844329834, |
| "eval_chemistry_mean_token_accuracy": 0.6105219907760621, |
| "eval_chemistry_num_tokens": 34128558.0, |
| "eval_chemistry_runtime": 61.7645, |
| "eval_chemistry_samples_per_second": 8.095, |
| "eval_chemistry_steps_per_second": 2.024, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_math_entropy": 0.8534087224006652, |
| "eval_math_loss": 0.8671056628227234, |
| "eval_math_mean_token_accuracy": 0.7843804006576538, |
| "eval_math_num_tokens": 34128558.0, |
| "eval_math_runtime": 63.3435, |
| "eval_math_samples_per_second": 7.893, |
| "eval_math_steps_per_second": 1.973, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_physics_entropy": 1.501321719646454, |
| "eval_physics_loss": 1.5970268249511719, |
| "eval_physics_mean_token_accuracy": 0.677275146484375, |
| "eval_physics_num_tokens": 34128558.0, |
| "eval_physics_runtime": 71.8682, |
| "eval_physics_samples_per_second": 6.957, |
| "eval_physics_steps_per_second": 1.739, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.8420938113704324, |
| "epoch": 1.936, |
| "grad_norm": 23.0, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 13.4465, |
| "mean_token_accuracy": 0.7893414869904518, |
| "num_tokens": 34408056.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.8260930391028524, |
| "epoch": 1.952, |
| "grad_norm": 20.25, |
| "learning_rate": 1.3192592592592594e-05, |
| "loss": 13.351, |
| "mean_token_accuracy": 0.7903814427554607, |
| "num_tokens": 34684679.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.854944153688848, |
| "epoch": 1.968, |
| "grad_norm": 20.25, |
| "learning_rate": 1.311851851851852e-05, |
| "loss": 13.8537, |
| "mean_token_accuracy": 0.7848830681294203, |
| "num_tokens": 34971038.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.8286283385008574, |
| "epoch": 1.984, |
| "grad_norm": 17.375, |
| "learning_rate": 1.3044444444444446e-05, |
| "loss": 13.2603, |
| "mean_token_accuracy": 0.792245727404952, |
| "num_tokens": 35262281.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.8108164651319385, |
| "epoch": 2.0, |
| "grad_norm": 23.5, |
| "learning_rate": 1.297037037037037e-05, |
| "loss": 13.0654, |
| "mean_token_accuracy": 0.7937818571925164, |
| "num_tokens": 35560864.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.8416191145777703, |
| "epoch": 2.016, |
| "grad_norm": 16.625, |
| "learning_rate": 1.2896296296296299e-05, |
| "loss": 13.4754, |
| "mean_token_accuracy": 0.7906782515347004, |
| "num_tokens": 35846704.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.8315626280382276, |
| "epoch": 2.032, |
| "grad_norm": 22.875, |
| "learning_rate": 1.2822222222222222e-05, |
| "loss": 13.3097, |
| "mean_token_accuracy": 0.7906810797750949, |
| "num_tokens": 36128775.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.8032516647130251, |
| "epoch": 2.048, |
| "grad_norm": 19.5, |
| "learning_rate": 1.274814814814815e-05, |
| "loss": 12.8687, |
| "mean_token_accuracy": 0.7970194958150387, |
| "num_tokens": 36419504.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.8010448306798935, |
| "epoch": 2.064, |
| "grad_norm": 18.625, |
| "learning_rate": 1.2674074074074075e-05, |
| "loss": 12.9604, |
| "mean_token_accuracy": 0.7955828506499529, |
| "num_tokens": 36706816.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.8162666315212845, |
| "epoch": 2.08, |
| "grad_norm": 18.875, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 12.9645, |
| "mean_token_accuracy": 0.7942825101315976, |
| "num_tokens": 36988475.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_biology_entropy": 3.279397897720337, |
| "eval_biology_loss": 3.8041577339172363, |
| "eval_biology_mean_token_accuracy": 0.42471638417243956, |
| "eval_biology_num_tokens": 36988475.0, |
| "eval_biology_runtime": 49.9068, |
| "eval_biology_samples_per_second": 10.019, |
| "eval_biology_steps_per_second": 2.505, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_chemistry_entropy": 1.8754352440834046, |
| "eval_chemistry_loss": 2.0649032592773438, |
| "eval_chemistry_mean_token_accuracy": 0.6106975488662719, |
| "eval_chemistry_num_tokens": 36988475.0, |
| "eval_chemistry_runtime": 62.0223, |
| "eval_chemistry_samples_per_second": 8.062, |
| "eval_chemistry_steps_per_second": 2.015, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_math_entropy": 0.8302458760738373, |
| "eval_math_loss": 0.8646599054336548, |
| "eval_math_mean_token_accuracy": 0.7845497117042541, |
| "eval_math_num_tokens": 36988475.0, |
| "eval_math_runtime": 63.2985, |
| "eval_math_samples_per_second": 7.899, |
| "eval_math_steps_per_second": 1.975, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_physics_entropy": 1.449456007003784, |
| "eval_physics_loss": 1.5842238664627075, |
| "eval_physics_mean_token_accuracy": 0.6789066867828369, |
| "eval_physics_num_tokens": 36988475.0, |
| "eval_physics_runtime": 71.9415, |
| "eval_physics_samples_per_second": 6.95, |
| "eval_physics_steps_per_second": 1.738, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.8240997634828091, |
| "epoch": 2.096, |
| "grad_norm": 18.625, |
| "learning_rate": 1.2525925925925928e-05, |
| "loss": 13.3182, |
| "mean_token_accuracy": 0.7931611612439156, |
| "num_tokens": 37270131.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.7984889261424541, |
| "epoch": 2.112, |
| "grad_norm": 17.125, |
| "learning_rate": 1.2451851851851853e-05, |
| "loss": 12.8484, |
| "mean_token_accuracy": 0.7954089492559433, |
| "num_tokens": 37563537.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.7992523608729243, |
| "epoch": 2.128, |
| "grad_norm": 21.25, |
| "learning_rate": 1.237777777777778e-05, |
| "loss": 12.7652, |
| "mean_token_accuracy": 0.7969741750508547, |
| "num_tokens": 37843959.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.8009057834744453, |
| "epoch": 2.144, |
| "grad_norm": 18.125, |
| "learning_rate": 1.2303703703703704e-05, |
| "loss": 12.9337, |
| "mean_token_accuracy": 0.7958127073943615, |
| "num_tokens": 38133092.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.8111963203176856, |
| "epoch": 2.16, |
| "grad_norm": 17.0, |
| "learning_rate": 1.222962962962963e-05, |
| "loss": 12.9893, |
| "mean_token_accuracy": 0.7958234615623951, |
| "num_tokens": 38421229.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.7792808337137103, |
| "epoch": 2.176, |
| "grad_norm": 21.0, |
| "learning_rate": 1.2155555555555555e-05, |
| "loss": 12.5213, |
| "mean_token_accuracy": 0.8006426334381104, |
| "num_tokens": 38708043.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.807457305304706, |
| "epoch": 2.192, |
| "grad_norm": 18.0, |
| "learning_rate": 1.2081481481481484e-05, |
| "loss": 12.9771, |
| "mean_token_accuracy": 0.7937675736844539, |
| "num_tokens": 38996930.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.8185192277655006, |
| "epoch": 2.208, |
| "grad_norm": 33.25, |
| "learning_rate": 1.2007407407407408e-05, |
| "loss": 13.2358, |
| "mean_token_accuracy": 0.7927701294422149, |
| "num_tokens": 39279481.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.8069099761545658, |
| "epoch": 2.224, |
| "grad_norm": 20.875, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 13.003, |
| "mean_token_accuracy": 0.7952508442103863, |
| "num_tokens": 39569030.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.8127697564661502, |
| "epoch": 2.24, |
| "grad_norm": 19.875, |
| "learning_rate": 1.185925925925926e-05, |
| "loss": 13.0199, |
| "mean_token_accuracy": 0.7938791792839766, |
| "num_tokens": 39854873.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_biology_entropy": 3.3652866687774656, |
| "eval_biology_loss": 3.8636908531188965, |
| "eval_biology_mean_token_accuracy": 0.42383125352859496, |
| "eval_biology_num_tokens": 39854873.0, |
| "eval_biology_runtime": 50.4898, |
| "eval_biology_samples_per_second": 9.903, |
| "eval_biology_steps_per_second": 2.476, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_chemistry_entropy": 1.9101995310783386, |
| "eval_chemistry_loss": 2.068533182144165, |
| "eval_chemistry_mean_token_accuracy": 0.6113230414390564, |
| "eval_chemistry_num_tokens": 39854873.0, |
| "eval_chemistry_runtime": 61.9928, |
| "eval_chemistry_samples_per_second": 8.065, |
| "eval_chemistry_steps_per_second": 2.016, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_math_entropy": 0.8333683574199676, |
| "eval_math_loss": 0.862160861492157, |
| "eval_math_mean_token_accuracy": 0.7853630857467652, |
| "eval_math_num_tokens": 39854873.0, |
| "eval_math_runtime": 63.6301, |
| "eval_math_samples_per_second": 7.858, |
| "eval_math_steps_per_second": 1.964, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_physics_entropy": 1.4711959066390992, |
| "eval_physics_loss": 1.5825717449188232, |
| "eval_physics_mean_token_accuracy": 0.6793005049228669, |
| "eval_physics_num_tokens": 39854873.0, |
| "eval_physics_runtime": 71.9736, |
| "eval_physics_samples_per_second": 6.947, |
| "eval_physics_steps_per_second": 1.737, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.8025357913225889, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 17.875, |
| "learning_rate": 1.1785185185185186e-05, |
| "loss": 12.9019, |
| "mean_token_accuracy": 0.7952509053051472, |
| "num_tokens": 40141190.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.8296092342585325, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 23.625, |
| "learning_rate": 1.1711111111111113e-05, |
| "loss": 13.3226, |
| "mean_token_accuracy": 0.7917913179844618, |
| "num_tokens": 40415203.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.8058724822476506, |
| "epoch": 2.288, |
| "grad_norm": 18.75, |
| "learning_rate": 1.1637037037037037e-05, |
| "loss": 12.8907, |
| "mean_token_accuracy": 0.7941552817821502, |
| "num_tokens": 40702393.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.8097540365532041, |
| "epoch": 2.304, |
| "grad_norm": 22.125, |
| "learning_rate": 1.1562962962962964e-05, |
| "loss": 12.9914, |
| "mean_token_accuracy": 0.7962609365582466, |
| "num_tokens": 40982775.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.818249425292015, |
| "epoch": 2.32, |
| "grad_norm": 21.625, |
| "learning_rate": 1.1488888888888889e-05, |
| "loss": 13.1437, |
| "mean_token_accuracy": 0.7941067516803741, |
| "num_tokens": 41263356.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.8201307155191898, |
| "epoch": 2.336, |
| "grad_norm": 19.125, |
| "learning_rate": 1.1414814814814817e-05, |
| "loss": 13.226, |
| "mean_token_accuracy": 0.7925650533288717, |
| "num_tokens": 41545235.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.8251900613307953, |
| "epoch": 2.352, |
| "grad_norm": 20.0, |
| "learning_rate": 1.1340740740740742e-05, |
| "loss": 13.2799, |
| "mean_token_accuracy": 0.7902849771082401, |
| "num_tokens": 41833417.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.8109830755740404, |
| "epoch": 2.368, |
| "grad_norm": 20.375, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 13.0504, |
| "mean_token_accuracy": 0.7949678536504508, |
| "num_tokens": 42117030.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.8299402013421059, |
| "epoch": 2.384, |
| "grad_norm": 24.75, |
| "learning_rate": 1.1192592592592593e-05, |
| "loss": 13.3754, |
| "mean_token_accuracy": 0.7900113243609667, |
| "num_tokens": 42410990.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.8205796327441931, |
| "epoch": 2.4, |
| "grad_norm": 21.5, |
| "learning_rate": 1.111851851851852e-05, |
| "loss": 13.1362, |
| "mean_token_accuracy": 0.7933749638497829, |
| "num_tokens": 42691890.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_biology_entropy": 3.2996106805801393, |
| "eval_biology_loss": 3.817183494567871, |
| "eval_biology_mean_token_accuracy": 0.42520643401145936, |
| "eval_biology_num_tokens": 42691890.0, |
| "eval_biology_runtime": 50.213, |
| "eval_biology_samples_per_second": 9.958, |
| "eval_biology_steps_per_second": 2.489, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_chemistry_entropy": 1.8643895120620728, |
| "eval_chemistry_loss": 2.0574374198913574, |
| "eval_chemistry_mean_token_accuracy": 0.6119919688701629, |
| "eval_chemistry_num_tokens": 42691890.0, |
| "eval_chemistry_runtime": 61.4954, |
| "eval_chemistry_samples_per_second": 8.131, |
| "eval_chemistry_steps_per_second": 2.033, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_math_entropy": 0.8184586379528046, |
| "eval_math_loss": 0.8592761158943176, |
| "eval_math_mean_token_accuracy": 0.7856973071098328, |
| "eval_math_num_tokens": 42691890.0, |
| "eval_math_runtime": 63.4025, |
| "eval_math_samples_per_second": 7.886, |
| "eval_math_steps_per_second": 1.972, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_physics_entropy": 1.4436475372314452, |
| "eval_physics_loss": 1.5851181745529175, |
| "eval_physics_mean_token_accuracy": 0.6786722614765167, |
| "eval_physics_num_tokens": 42691890.0, |
| "eval_physics_runtime": 71.9274, |
| "eval_physics_samples_per_second": 6.951, |
| "eval_physics_steps_per_second": 1.738, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.8286989040672779, |
| "epoch": 2.416, |
| "grad_norm": 20.875, |
| "learning_rate": 1.1044444444444444e-05, |
| "loss": 13.3943, |
| "mean_token_accuracy": 0.7892198380082845, |
| "num_tokens": 42971588.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.8104729436337947, |
| "epoch": 2.432, |
| "grad_norm": 21.0, |
| "learning_rate": 1.0970370370370371e-05, |
| "loss": 12.9644, |
| "mean_token_accuracy": 0.7956920482218266, |
| "num_tokens": 43253821.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.7847399104386568, |
| "epoch": 2.448, |
| "grad_norm": 19.75, |
| "learning_rate": 1.0896296296296298e-05, |
| "loss": 12.7387, |
| "mean_token_accuracy": 0.7986189529299736, |
| "num_tokens": 43550902.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.8245068648830056, |
| "epoch": 2.464, |
| "grad_norm": 19.0, |
| "learning_rate": 1.0822222222222222e-05, |
| "loss": 13.1098, |
| "mean_token_accuracy": 0.7924448147416114, |
| "num_tokens": 43844259.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.863299522548914, |
| "epoch": 2.48, |
| "grad_norm": 32.0, |
| "learning_rate": 1.074814814814815e-05, |
| "loss": 13.8505, |
| "mean_token_accuracy": 0.7850053083151579, |
| "num_tokens": 44115701.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.8298073586076498, |
| "epoch": 2.496, |
| "grad_norm": 18.125, |
| "learning_rate": 1.0674074074074074e-05, |
| "loss": 13.4252, |
| "mean_token_accuracy": 0.7907994110137224, |
| "num_tokens": 44405520.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.8181227730587125, |
| "epoch": 2.512, |
| "grad_norm": 18.125, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 13.0696, |
| "mean_token_accuracy": 0.7937217362225055, |
| "num_tokens": 44686477.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.814422365091741, |
| "epoch": 2.528, |
| "grad_norm": 18.625, |
| "learning_rate": 1.0525925925925927e-05, |
| "loss": 13.1102, |
| "mean_token_accuracy": 0.7930266965180636, |
| "num_tokens": 44969760.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.8113391397520899, |
| "epoch": 2.544, |
| "grad_norm": 20.625, |
| "learning_rate": 1.0451851851851853e-05, |
| "loss": 13.0414, |
| "mean_token_accuracy": 0.7953265462070703, |
| "num_tokens": 45255326.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.8113772097975016, |
| "epoch": 2.56, |
| "grad_norm": 19.125, |
| "learning_rate": 1.0377777777777778e-05, |
| "loss": 13.0064, |
| "mean_token_accuracy": 0.7931508217006922, |
| "num_tokens": 45532525.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_biology_entropy": 3.3154334926605227, |
| "eval_biology_loss": 3.8110804557800293, |
| "eval_biology_mean_token_accuracy": 0.4255426707267761, |
| "eval_biology_num_tokens": 45532525.0, |
| "eval_biology_runtime": 49.6283, |
| "eval_biology_samples_per_second": 10.075, |
| "eval_biology_steps_per_second": 2.519, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_chemistry_entropy": 1.8724517908096314, |
| "eval_chemistry_loss": 2.058601140975952, |
| "eval_chemistry_mean_token_accuracy": 0.6117340788841248, |
| "eval_chemistry_num_tokens": 45532525.0, |
| "eval_chemistry_runtime": 61.5463, |
| "eval_chemistry_samples_per_second": 8.124, |
| "eval_chemistry_steps_per_second": 2.031, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_math_entropy": 0.8175128948688507, |
| "eval_math_loss": 0.8564888834953308, |
| "eval_math_mean_token_accuracy": 0.7863039264678955, |
| "eval_math_num_tokens": 45532525.0, |
| "eval_math_runtime": 63.614, |
| "eval_math_samples_per_second": 7.86, |
| "eval_math_steps_per_second": 1.965, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_physics_entropy": 1.4472115359306335, |
| "eval_physics_loss": 1.5836905241012573, |
| "eval_physics_mean_token_accuracy": 0.6789892053604126, |
| "eval_physics_num_tokens": 45532525.0, |
| "eval_physics_runtime": 71.9282, |
| "eval_physics_samples_per_second": 6.951, |
| "eval_physics_steps_per_second": 1.738, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.8044116089120508, |
| "epoch": 2.576, |
| "grad_norm": 19.25, |
| "learning_rate": 1.0303703703703705e-05, |
| "loss": 13.0547, |
| "mean_token_accuracy": 0.7951739374548197, |
| "num_tokens": 45817478.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.7988964702934027, |
| "epoch": 2.592, |
| "grad_norm": 27.875, |
| "learning_rate": 1.0229629629629631e-05, |
| "loss": 12.8268, |
| "mean_token_accuracy": 0.7968171291053295, |
| "num_tokens": 46109575.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.8271580126136542, |
| "epoch": 2.608, |
| "grad_norm": 23.0, |
| "learning_rate": 1.0155555555555556e-05, |
| "loss": 13.3287, |
| "mean_token_accuracy": 0.7899187684059144, |
| "num_tokens": 46391461.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.7868071464821697, |
| "epoch": 2.624, |
| "grad_norm": 19.375, |
| "learning_rate": 1.0081481481481484e-05, |
| "loss": 12.6798, |
| "mean_token_accuracy": 0.8014541812241077, |
| "num_tokens": 46683117.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.8248436784371733, |
| "epoch": 2.64, |
| "grad_norm": 20.125, |
| "learning_rate": 1.0007407407407407e-05, |
| "loss": 13.1653, |
| "mean_token_accuracy": 0.7917656779289246, |
| "num_tokens": 46965534.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.7827718367800116, |
| "epoch": 2.656, |
| "grad_norm": 20.25, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 12.5768, |
| "mean_token_accuracy": 0.79845260232687, |
| "num_tokens": 47245852.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.8232856251299381, |
| "epoch": 2.672, |
| "grad_norm": 27.5, |
| "learning_rate": 9.85925925925926e-06, |
| "loss": 13.2773, |
| "mean_token_accuracy": 0.7893969945609569, |
| "num_tokens": 47524916.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.8051718821749091, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 19.5, |
| "learning_rate": 9.785185185185187e-06, |
| "loss": 12.9785, |
| "mean_token_accuracy": 0.7945211797952652, |
| "num_tokens": 47807131.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.8211181791499257, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 22.25, |
| "learning_rate": 9.711111111111111e-06, |
| "loss": 13.1871, |
| "mean_token_accuracy": 0.7917978953570127, |
| "num_tokens": 48099654.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.8078500226140022, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 19.625, |
| "learning_rate": 9.637037037037038e-06, |
| "loss": 13.0211, |
| "mean_token_accuracy": 0.7936208248138428, |
| "num_tokens": 48375019.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_biology_entropy": 3.338820819854736, |
| "eval_biology_loss": 3.842472553253174, |
| "eval_biology_mean_token_accuracy": 0.4245969302654266, |
| "eval_biology_num_tokens": 48375019.0, |
| "eval_biology_runtime": 49.7408, |
| "eval_biology_samples_per_second": 10.052, |
| "eval_biology_steps_per_second": 2.513, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_chemistry_entropy": 1.8794775671958923, |
| "eval_chemistry_loss": 2.056818962097168, |
| "eval_chemistry_mean_token_accuracy": 0.6125242853164673, |
| "eval_chemistry_num_tokens": 48375019.0, |
| "eval_chemistry_runtime": 63.286, |
| "eval_chemistry_samples_per_second": 7.901, |
| "eval_chemistry_steps_per_second": 1.975, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_math_entropy": 0.8294126708507538, |
| "eval_math_loss": 0.8545240759849548, |
| "eval_math_mean_token_accuracy": 0.7866081275939941, |
| "eval_math_num_tokens": 48375019.0, |
| "eval_math_runtime": 64.1496, |
| "eval_math_samples_per_second": 7.794, |
| "eval_math_steps_per_second": 1.949, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_physics_entropy": 1.4557471733093261, |
| "eval_physics_loss": 1.5759708881378174, |
| "eval_physics_mean_token_accuracy": 0.6801469361782074, |
| "eval_physics_num_tokens": 48375019.0, |
| "eval_physics_runtime": 72.8048, |
| "eval_physics_samples_per_second": 6.868, |
| "eval_physics_steps_per_second": 1.717, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.8175020966678858, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 20.5, |
| "learning_rate": 9.562962962962965e-06, |
| "loss": 13.0885, |
| "mean_token_accuracy": 0.7923605650663376, |
| "num_tokens": 48659284.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.7984929440543056, |
| "epoch": 2.752, |
| "grad_norm": 19.0, |
| "learning_rate": 9.48888888888889e-06, |
| "loss": 12.928, |
| "mean_token_accuracy": 0.7956108339130878, |
| "num_tokens": 48943804.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.8201850002631546, |
| "epoch": 2.768, |
| "grad_norm": 19.75, |
| "learning_rate": 9.414814814814816e-06, |
| "loss": 13.0841, |
| "mean_token_accuracy": 0.7943447981029749, |
| "num_tokens": 49230939.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.827330220490694, |
| "epoch": 2.784, |
| "grad_norm": 21.25, |
| "learning_rate": 9.34074074074074e-06, |
| "loss": 13.4156, |
| "mean_token_accuracy": 0.7888516779989004, |
| "num_tokens": 49504425.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.8066693387925625, |
| "epoch": 2.8, |
| "grad_norm": 16.5, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 13.0824, |
| "mean_token_accuracy": 0.7948512561619282, |
| "num_tokens": 49782661.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.819707078486681, |
| "epoch": 2.816, |
| "grad_norm": 24.0, |
| "learning_rate": 9.192592592592594e-06, |
| "loss": 13.1454, |
| "mean_token_accuracy": 0.79271629601717, |
| "num_tokens": 50073632.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.8052127622067928, |
| "epoch": 2.832, |
| "grad_norm": 20.5, |
| "learning_rate": 9.118518518518518e-06, |
| "loss": 12.9015, |
| "mean_token_accuracy": 0.7952382136136293, |
| "num_tokens": 50356964.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.8257588528096675, |
| "epoch": 2.848, |
| "grad_norm": 18.0, |
| "learning_rate": 9.044444444444445e-06, |
| "loss": 13.3851, |
| "mean_token_accuracy": 0.7918277770280838, |
| "num_tokens": 50644535.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.8250102279707789, |
| "epoch": 2.864, |
| "grad_norm": 23.5, |
| "learning_rate": 8.970370370370372e-06, |
| "loss": 13.2005, |
| "mean_token_accuracy": 0.7922936867922544, |
| "num_tokens": 50925653.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.820320550724864, |
| "epoch": 2.88, |
| "grad_norm": 20.625, |
| "learning_rate": 8.896296296296298e-06, |
| "loss": 13.0627, |
| "mean_token_accuracy": 0.7933201365172863, |
| "num_tokens": 51204374.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_biology_entropy": 3.31503232383728, |
| "eval_biology_loss": 3.8447117805480957, |
| "eval_biology_mean_token_accuracy": 0.4249435958862305, |
| "eval_biology_num_tokens": 51204374.0, |
| "eval_biology_runtime": 49.695, |
| "eval_biology_samples_per_second": 10.061, |
| "eval_biology_steps_per_second": 2.515, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_chemistry_entropy": 1.864605971813202, |
| "eval_chemistry_loss": 2.0581791400909424, |
| "eval_chemistry_mean_token_accuracy": 0.6130386800765991, |
| "eval_chemistry_num_tokens": 51204374.0, |
| "eval_chemistry_runtime": 61.9044, |
| "eval_chemistry_samples_per_second": 8.077, |
| "eval_chemistry_steps_per_second": 2.019, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_math_entropy": 0.8211403098106385, |
| "eval_math_loss": 0.8531603813171387, |
| "eval_math_mean_token_accuracy": 0.7870454635620118, |
| "eval_math_num_tokens": 51204374.0, |
| "eval_math_runtime": 63.6675, |
| "eval_math_samples_per_second": 7.853, |
| "eval_math_steps_per_second": 1.963, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_physics_entropy": 1.4479440732002258, |
| "eval_physics_loss": 1.5755829811096191, |
| "eval_physics_mean_token_accuracy": 0.6802050042152404, |
| "eval_physics_num_tokens": 51204374.0, |
| "eval_physics_runtime": 72.2085, |
| "eval_physics_samples_per_second": 6.924, |
| "eval_physics_steps_per_second": 1.731, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.7931723792105914, |
| "epoch": 2.896, |
| "grad_norm": 22.25, |
| "learning_rate": 8.822222222222223e-06, |
| "loss": 12.8232, |
| "mean_token_accuracy": 0.7966993600130081, |
| "num_tokens": 51483944.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.8129935130476952, |
| "epoch": 2.912, |
| "grad_norm": 17.625, |
| "learning_rate": 8.74814814814815e-06, |
| "loss": 13.1195, |
| "mean_token_accuracy": 0.794299291819334, |
| "num_tokens": 51765755.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.8070885950699449, |
| "epoch": 2.928, |
| "grad_norm": 21.5, |
| "learning_rate": 8.674074074074074e-06, |
| "loss": 12.9814, |
| "mean_token_accuracy": 0.7966341748833656, |
| "num_tokens": 52056379.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.798726899549365, |
| "epoch": 2.944, |
| "grad_norm": 18.125, |
| "learning_rate": 8.6e-06, |
| "loss": 12.7059, |
| "mean_token_accuracy": 0.7972525604069233, |
| "num_tokens": 52346232.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.8092642085626721, |
| "epoch": 2.96, |
| "grad_norm": 17.625, |
| "learning_rate": 8.525925925925927e-06, |
| "loss": 13.0557, |
| "mean_token_accuracy": 0.7941314589232207, |
| "num_tokens": 52633789.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.7947347335517406, |
| "epoch": 2.976, |
| "grad_norm": 17.375, |
| "learning_rate": 8.451851851851852e-06, |
| "loss": 12.8431, |
| "mean_token_accuracy": 0.795494619384408, |
| "num_tokens": 52911755.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.804375764541328, |
| "epoch": 2.992, |
| "grad_norm": 20.375, |
| "learning_rate": 8.377777777777779e-06, |
| "loss": 12.8876, |
| "mean_token_accuracy": 0.7952792067080736, |
| "num_tokens": 53198176.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.8181121354922652, |
| "epoch": 3.008, |
| "grad_norm": 18.375, |
| "learning_rate": 8.303703703703705e-06, |
| "loss": 13.1365, |
| "mean_token_accuracy": 0.7920746240764857, |
| "num_tokens": 53481893.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.8169939275830984, |
| "epoch": 3.024, |
| "grad_norm": 18.375, |
| "learning_rate": 8.229629629629632e-06, |
| "loss": 13.0443, |
| "mean_token_accuracy": 0.7934746831655503, |
| "num_tokens": 53771717.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.8134762601926923, |
| "epoch": 3.04, |
| "grad_norm": 19.125, |
| "learning_rate": 8.155555555555556e-06, |
| "loss": 13.067, |
| "mean_token_accuracy": 0.7942324243485928, |
| "num_tokens": 54058045.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_biology_entropy": 3.291802607536316, |
| "eval_biology_loss": 3.8235349655151367, |
| "eval_biology_mean_token_accuracy": 0.4258481192588806, |
| "eval_biology_num_tokens": 54058045.0, |
| "eval_biology_runtime": 49.6604, |
| "eval_biology_samples_per_second": 10.068, |
| "eval_biology_steps_per_second": 2.517, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_chemistry_entropy": 1.8494226307868957, |
| "eval_chemistry_loss": 2.0516436100006104, |
| "eval_chemistry_mean_token_accuracy": 0.613143976688385, |
| "eval_chemistry_num_tokens": 54058045.0, |
| "eval_chemistry_runtime": 61.6855, |
| "eval_chemistry_samples_per_second": 8.106, |
| "eval_chemistry_steps_per_second": 2.026, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_math_entropy": 0.8140118596553803, |
| "eval_math_loss": 0.8527244925498962, |
| "eval_math_mean_token_accuracy": 0.7870401821136475, |
| "eval_math_num_tokens": 54058045.0, |
| "eval_math_runtime": 63.2251, |
| "eval_math_samples_per_second": 7.908, |
| "eval_math_steps_per_second": 1.977, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_physics_entropy": 1.4363980693817138, |
| "eval_physics_loss": 1.574827790260315, |
| "eval_physics_mean_token_accuracy": 0.6798698828220368, |
| "eval_physics_num_tokens": 54058045.0, |
| "eval_physics_runtime": 72.2912, |
| "eval_physics_samples_per_second": 6.916, |
| "eval_physics_steps_per_second": 1.729, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.8106893755495548, |
| "epoch": 3.056, |
| "grad_norm": 16.875, |
| "learning_rate": 8.081481481481483e-06, |
| "loss": 13.0624, |
| "mean_token_accuracy": 0.7939662571996451, |
| "num_tokens": 54334332.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.7890156630426646, |
| "epoch": 3.072, |
| "grad_norm": 18.125, |
| "learning_rate": 8.007407407407408e-06, |
| "loss": 12.7017, |
| "mean_token_accuracy": 0.7977974370121956, |
| "num_tokens": 54624543.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.7826174193993211, |
| "epoch": 3.088, |
| "grad_norm": 18.875, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 12.5541, |
| "mean_token_accuracy": 0.8012233301997185, |
| "num_tokens": 54907550.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.7943657983094454, |
| "epoch": 3.104, |
| "grad_norm": 16.375, |
| "learning_rate": 7.859259259259259e-06, |
| "loss": 12.7072, |
| "mean_token_accuracy": 0.7982670705765486, |
| "num_tokens": 55190959.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.7860442865639925, |
| "epoch": 3.12, |
| "grad_norm": 19.875, |
| "learning_rate": 7.785185185185185e-06, |
| "loss": 12.5807, |
| "mean_token_accuracy": 0.8005135275423527, |
| "num_tokens": 55481635.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.7800285588949919, |
| "epoch": 3.136, |
| "grad_norm": 19.125, |
| "learning_rate": 7.711111111111112e-06, |
| "loss": 12.6268, |
| "mean_token_accuracy": 0.7990261886268855, |
| "num_tokens": 55769010.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.8018075853586197, |
| "epoch": 3.152, |
| "grad_norm": 19.25, |
| "learning_rate": 7.637037037037037e-06, |
| "loss": 12.8788, |
| "mean_token_accuracy": 0.7970602057874203, |
| "num_tokens": 56053160.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.8143207451328636, |
| "epoch": 3.168, |
| "grad_norm": 18.125, |
| "learning_rate": 7.562962962962963e-06, |
| "loss": 13.0717, |
| "mean_token_accuracy": 0.7944711521267891, |
| "num_tokens": 56337066.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.7893068237230182, |
| "epoch": 3.184, |
| "grad_norm": 21.0, |
| "learning_rate": 7.48888888888889e-06, |
| "loss": 12.6427, |
| "mean_token_accuracy": 0.7965629179030657, |
| "num_tokens": 56618899.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.7735519032925368, |
| "epoch": 3.2, |
| "grad_norm": 15.1875, |
| "learning_rate": 7.4148148148148155e-06, |
| "loss": 12.4168, |
| "mean_token_accuracy": 0.8018718279898167, |
| "num_tokens": 56910071.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_biology_entropy": 3.2790002565383913, |
| "eval_biology_loss": 3.8444063663482666, |
| "eval_biology_mean_token_accuracy": 0.4248851606845856, |
| "eval_biology_num_tokens": 56910071.0, |
| "eval_biology_runtime": 49.572, |
| "eval_biology_samples_per_second": 10.086, |
| "eval_biology_steps_per_second": 2.522, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_chemistry_entropy": 1.8469837627410888, |
| "eval_chemistry_loss": 2.0660977363586426, |
| "eval_chemistry_mean_token_accuracy": 0.6124599766731262, |
| "eval_chemistry_num_tokens": 56910071.0, |
| "eval_chemistry_runtime": 60.9402, |
| "eval_chemistry_samples_per_second": 8.205, |
| "eval_chemistry_steps_per_second": 2.051, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_math_entropy": 0.8110623922348023, |
| "eval_math_loss": 0.8516995310783386, |
| "eval_math_mean_token_accuracy": 0.7868917775154114, |
| "eval_math_num_tokens": 56910071.0, |
| "eval_math_runtime": 62.6639, |
| "eval_math_samples_per_second": 7.979, |
| "eval_math_steps_per_second": 1.995, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_physics_entropy": 1.4288099465370179, |
| "eval_physics_loss": 1.5777822732925415, |
| "eval_physics_mean_token_accuracy": 0.6803083395957947, |
| "eval_physics_num_tokens": 56910071.0, |
| "eval_physics_runtime": 71.6833, |
| "eval_physics_samples_per_second": 6.975, |
| "eval_physics_steps_per_second": 1.744, |
| "step": 2000 |
| }, |
| { |
| "entropy": 0.8168226035311819, |
| "epoch": 3.216, |
| "grad_norm": 21.125, |
| "learning_rate": 7.340740740740742e-06, |
| "loss": 13.2201, |
| "mean_token_accuracy": 0.7916460353881121, |
| "num_tokens": 57190265.0, |
| "step": 2010 |
| }, |
| { |
| "entropy": 0.8096309691667557, |
| "epoch": 3.232, |
| "grad_norm": 20.125, |
| "learning_rate": 7.266666666666668e-06, |
| "loss": 13.0224, |
| "mean_token_accuracy": 0.7954025827348232, |
| "num_tokens": 57471355.0, |
| "step": 2020 |
| }, |
| { |
| "entropy": 0.7891311341896653, |
| "epoch": 3.248, |
| "grad_norm": 24.0, |
| "learning_rate": 7.192592592592593e-06, |
| "loss": 12.5847, |
| "mean_token_accuracy": 0.7999192129820585, |
| "num_tokens": 57749030.0, |
| "step": 2030 |
| }, |
| { |
| "entropy": 0.7859420148655772, |
| "epoch": 3.2640000000000002, |
| "grad_norm": 19.5, |
| "learning_rate": 7.118518518518519e-06, |
| "loss": 12.5973, |
| "mean_token_accuracy": 0.7988406583666802, |
| "num_tokens": 58027567.0, |
| "step": 2040 |
| }, |
| { |
| "entropy": 0.8036651344969868, |
| "epoch": 3.2800000000000002, |
| "grad_norm": 16.375, |
| "learning_rate": 7.044444444444445e-06, |
| "loss": 13.0061, |
| "mean_token_accuracy": 0.7934975892305374, |
| "num_tokens": 58304346.0, |
| "step": 2050 |
| }, |
| { |
| "entropy": 0.8267201952636242, |
| "epoch": 3.296, |
| "grad_norm": 19.375, |
| "learning_rate": 6.97037037037037e-06, |
| "loss": 13.2243, |
| "mean_token_accuracy": 0.7927972044795751, |
| "num_tokens": 58579831.0, |
| "step": 2060 |
| }, |
| { |
| "entropy": 0.8057495908811688, |
| "epoch": 3.312, |
| "grad_norm": 19.75, |
| "learning_rate": 6.896296296296297e-06, |
| "loss": 12.9769, |
| "mean_token_accuracy": 0.7945728350430727, |
| "num_tokens": 58858426.0, |
| "step": 2070 |
| }, |
| { |
| "entropy": 0.7923213245347143, |
| "epoch": 3.328, |
| "grad_norm": 20.125, |
| "learning_rate": 6.8222222222222225e-06, |
| "loss": 12.6523, |
| "mean_token_accuracy": 0.7971304733306169, |
| "num_tokens": 59142949.0, |
| "step": 2080 |
| }, |
| { |
| "entropy": 0.7926076017320156, |
| "epoch": 3.344, |
| "grad_norm": 21.625, |
| "learning_rate": 6.748148148148149e-06, |
| "loss": 12.7328, |
| "mean_token_accuracy": 0.7980255495756865, |
| "num_tokens": 59424654.0, |
| "step": 2090 |
| }, |
| { |
| "entropy": 0.8348858727142214, |
| "epoch": 3.36, |
| "grad_norm": 21.875, |
| "learning_rate": 6.674074074074075e-06, |
| "loss": 13.6072, |
| "mean_token_accuracy": 0.7873939543962478, |
| "num_tokens": 59704876.0, |
| "step": 2100 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_biology_entropy": 3.2945481452941894, |
| "eval_biology_loss": 3.8381760120391846, |
| "eval_biology_mean_token_accuracy": 0.4253296444416046, |
| "eval_biology_num_tokens": 59704876.0, |
| "eval_biology_runtime": 49.6684, |
| "eval_biology_samples_per_second": 10.067, |
| "eval_biology_steps_per_second": 2.517, |
| "step": 2100 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_chemistry_entropy": 1.852885506629944, |
| "eval_chemistry_loss": 2.056868314743042, |
| "eval_chemistry_mean_token_accuracy": 0.6130528056621551, |
| "eval_chemistry_num_tokens": 59704876.0, |
| "eval_chemistry_runtime": 61.5842, |
| "eval_chemistry_samples_per_second": 8.119, |
| "eval_chemistry_steps_per_second": 2.03, |
| "step": 2100 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_math_entropy": 0.8173734290599823, |
| "eval_math_loss": 0.8512579202651978, |
| "eval_math_mean_token_accuracy": 0.7875010733604431, |
| "eval_math_num_tokens": 59704876.0, |
| "eval_math_runtime": 63.5404, |
| "eval_math_samples_per_second": 7.869, |
| "eval_math_steps_per_second": 1.967, |
| "step": 2100 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_physics_entropy": 1.4352242302894593, |
| "eval_physics_loss": 1.576236605644226, |
| "eval_physics_mean_token_accuracy": 0.6803803541660309, |
| "eval_physics_num_tokens": 59704876.0, |
| "eval_physics_runtime": 72.4573, |
| "eval_physics_samples_per_second": 6.901, |
| "eval_physics_steps_per_second": 1.725, |
| "step": 2100 |
| }, |
| { |
| "entropy": 0.8270608499646187, |
| "epoch": 3.376, |
| "grad_norm": 19.875, |
| "learning_rate": 6.600000000000001e-06, |
| "loss": 13.2402, |
| "mean_token_accuracy": 0.7916862409561872, |
| "num_tokens": 59988211.0, |
| "step": 2110 |
| }, |
| { |
| "entropy": 0.7938789419829846, |
| "epoch": 3.392, |
| "grad_norm": 17.875, |
| "learning_rate": 6.525925925925927e-06, |
| "loss": 12.632, |
| "mean_token_accuracy": 0.7999833203852177, |
| "num_tokens": 60276606.0, |
| "step": 2120 |
| }, |
| { |
| "entropy": 0.7785499131307005, |
| "epoch": 3.408, |
| "grad_norm": 16.625, |
| "learning_rate": 6.4518518518518525e-06, |
| "loss": 12.4635, |
| "mean_token_accuracy": 0.8002612300217151, |
| "num_tokens": 60560228.0, |
| "step": 2130 |
| }, |
| { |
| "entropy": 0.7914116440340877, |
| "epoch": 3.424, |
| "grad_norm": 20.25, |
| "learning_rate": 6.377777777777778e-06, |
| "loss": 12.7959, |
| "mean_token_accuracy": 0.7960323005914688, |
| "num_tokens": 60845893.0, |
| "step": 2140 |
| }, |
| { |
| "entropy": 0.7831454169005155, |
| "epoch": 3.44, |
| "grad_norm": 19.125, |
| "learning_rate": 6.303703703703704e-06, |
| "loss": 12.6096, |
| "mean_token_accuracy": 0.8000591345131397, |
| "num_tokens": 61126410.0, |
| "step": 2150 |
| }, |
| { |
| "entropy": 0.8127947147935629, |
| "epoch": 3.456, |
| "grad_norm": 20.625, |
| "learning_rate": 6.2296296296296295e-06, |
| "loss": 13.052, |
| "mean_token_accuracy": 0.7942486621439457, |
| "num_tokens": 61406501.0, |
| "step": 2160 |
| }, |
| { |
| "entropy": 0.8195335539057851, |
| "epoch": 3.472, |
| "grad_norm": 22.75, |
| "learning_rate": 6.155555555555556e-06, |
| "loss": 13.273, |
| "mean_token_accuracy": 0.7915166169404984, |
| "num_tokens": 61693412.0, |
| "step": 2170 |
| }, |
| { |
| "entropy": 0.7960003647953272, |
| "epoch": 3.488, |
| "grad_norm": 18.875, |
| "learning_rate": 6.081481481481482e-06, |
| "loss": 12.655, |
| "mean_token_accuracy": 0.7983210921287537, |
| "num_tokens": 61977303.0, |
| "step": 2180 |
| }, |
| { |
| "entropy": 0.7848934635519982, |
| "epoch": 3.504, |
| "grad_norm": 20.625, |
| "learning_rate": 6.007407407407407e-06, |
| "loss": 12.7621, |
| "mean_token_accuracy": 0.7969312839210033, |
| "num_tokens": 62265256.0, |
| "step": 2190 |
| }, |
| { |
| "entropy": 0.7879944819957018, |
| "epoch": 3.52, |
| "grad_norm": 21.125, |
| "learning_rate": 5.933333333333335e-06, |
| "loss": 12.6593, |
| "mean_token_accuracy": 0.798987215384841, |
| "num_tokens": 62554970.0, |
| "step": 2200 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_biology_entropy": 3.2846322717666627, |
| "eval_biology_loss": 3.8201656341552734, |
| "eval_biology_mean_token_accuracy": 0.4255785467624664, |
| "eval_biology_num_tokens": 62554970.0, |
| "eval_biology_runtime": 49.6539, |
| "eval_biology_samples_per_second": 10.07, |
| "eval_biology_steps_per_second": 2.517, |
| "step": 2200 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_chemistry_entropy": 1.844938889503479, |
| "eval_chemistry_loss": 2.0524256229400635, |
| "eval_chemistry_mean_token_accuracy": 0.6133026978969573, |
| "eval_chemistry_num_tokens": 62554970.0, |
| "eval_chemistry_runtime": 61.5813, |
| "eval_chemistry_samples_per_second": 8.119, |
| "eval_chemistry_steps_per_second": 2.03, |
| "step": 2200 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_math_entropy": 0.8196575653553009, |
| "eval_math_loss": 0.8505458831787109, |
| "eval_math_mean_token_accuracy": 0.7872078022956848, |
| "eval_math_num_tokens": 62554970.0, |
| "eval_math_runtime": 63.1336, |
| "eval_math_samples_per_second": 7.92, |
| "eval_math_steps_per_second": 1.98, |
| "step": 2200 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_physics_entropy": 1.4334162487983704, |
| "eval_physics_loss": 1.575282096862793, |
| "eval_physics_mean_token_accuracy": 0.6802379956245422, |
| "eval_physics_num_tokens": 62554970.0, |
| "eval_physics_runtime": 72.8294, |
| "eval_physics_samples_per_second": 6.865, |
| "eval_physics_steps_per_second": 1.716, |
| "step": 2200 |
| }, |
| { |
| "entropy": 0.8085380343720316, |
| "epoch": 3.536, |
| "grad_norm": 20.25, |
| "learning_rate": 5.85925925925926e-06, |
| "loss": 12.9996, |
| "mean_token_accuracy": 0.7948060061782598, |
| "num_tokens": 62837055.0, |
| "step": 2210 |
| }, |
| { |
| "entropy": 0.7764543682336807, |
| "epoch": 3.552, |
| "grad_norm": 18.25, |
| "learning_rate": 5.785185185185186e-06, |
| "loss": 12.4785, |
| "mean_token_accuracy": 0.8001452952623367, |
| "num_tokens": 63129767.0, |
| "step": 2220 |
| }, |
| { |
| "entropy": 0.7797821855172515, |
| "epoch": 3.568, |
| "grad_norm": 21.0, |
| "learning_rate": 5.711111111111112e-06, |
| "loss": 12.4244, |
| "mean_token_accuracy": 0.8012934874743223, |
| "num_tokens": 63424239.0, |
| "step": 2230 |
| }, |
| { |
| "entropy": 0.8088451908901334, |
| "epoch": 3.584, |
| "grad_norm": 23.375, |
| "learning_rate": 5.637037037037037e-06, |
| "loss": 12.984, |
| "mean_token_accuracy": 0.7959229256957769, |
| "num_tokens": 63704243.0, |
| "step": 2240 |
| }, |
| { |
| "entropy": 0.7592525988817215, |
| "epoch": 3.6, |
| "grad_norm": 18.875, |
| "learning_rate": 5.562962962962963e-06, |
| "loss": 12.2461, |
| "mean_token_accuracy": 0.8040501173585654, |
| "num_tokens": 63994604.0, |
| "step": 2250 |
| }, |
| { |
| "entropy": 0.8018990097567439, |
| "epoch": 3.616, |
| "grad_norm": 23.625, |
| "learning_rate": 5.4888888888888895e-06, |
| "loss": 12.9618, |
| "mean_token_accuracy": 0.7943440668284893, |
| "num_tokens": 64280032.0, |
| "step": 2260 |
| }, |
| { |
| "entropy": 0.7837540531530977, |
| "epoch": 3.632, |
| "grad_norm": 19.75, |
| "learning_rate": 5.414814814814815e-06, |
| "loss": 12.6335, |
| "mean_token_accuracy": 0.7986758667975664, |
| "num_tokens": 64560379.0, |
| "step": 2270 |
| }, |
| { |
| "entropy": 0.7692544776946306, |
| "epoch": 3.648, |
| "grad_norm": 21.25, |
| "learning_rate": 5.340740740740741e-06, |
| "loss": 12.3668, |
| "mean_token_accuracy": 0.8036178927868605, |
| "num_tokens": 64843196.0, |
| "step": 2280 |
| }, |
| { |
| "entropy": 0.8227372044697404, |
| "epoch": 3.664, |
| "grad_norm": 19.75, |
| "learning_rate": 5.2666666666666665e-06, |
| "loss": 13.246, |
| "mean_token_accuracy": 0.7918700549751521, |
| "num_tokens": 65126428.0, |
| "step": 2290 |
| }, |
| { |
| "entropy": 0.7894239040091634, |
| "epoch": 3.68, |
| "grad_norm": 22.125, |
| "learning_rate": 5.192592592592594e-06, |
| "loss": 12.6312, |
| "mean_token_accuracy": 0.7994260579347611, |
| "num_tokens": 65411914.0, |
| "step": 2300 |
| }, |
| { |
| "epoch": 3.68, |
| "eval_biology_entropy": 3.2819386901855467, |
| "eval_biology_loss": 3.8254449367523193, |
| "eval_biology_mean_token_accuracy": 0.42563737988471984, |
| "eval_biology_num_tokens": 65411914.0, |
| "eval_biology_runtime": 49.6604, |
| "eval_biology_samples_per_second": 10.068, |
| "eval_biology_steps_per_second": 2.517, |
| "step": 2300 |
| }, |
| { |
| "epoch": 3.68, |
| "eval_chemistry_entropy": 1.8424931778907776, |
| "eval_chemistry_loss": 2.0471694469451904, |
| "eval_chemistry_mean_token_accuracy": 0.6136951448917389, |
| "eval_chemistry_num_tokens": 65411914.0, |
| "eval_chemistry_runtime": 61.6168, |
| "eval_chemistry_samples_per_second": 8.115, |
| "eval_chemistry_steps_per_second": 2.029, |
| "step": 2300 |
| }, |
| { |
| "epoch": 3.68, |
| "eval_math_entropy": 0.8161297690868378, |
| "eval_math_loss": 0.8500486016273499, |
| "eval_math_mean_token_accuracy": 0.787582727432251, |
| "eval_math_num_tokens": 65411914.0, |
| "eval_math_runtime": 63.1651, |
| "eval_math_samples_per_second": 7.916, |
| "eval_math_steps_per_second": 1.979, |
| "step": 2300 |
| }, |
| { |
| "epoch": 3.68, |
| "eval_physics_entropy": 1.4325386419296264, |
| "eval_physics_loss": 1.5712970495224, |
| "eval_physics_mean_token_accuracy": 0.6806518092155457, |
| "eval_physics_num_tokens": 65411914.0, |
| "eval_physics_runtime": 72.2666, |
| "eval_physics_samples_per_second": 6.919, |
| "eval_physics_steps_per_second": 1.73, |
| "step": 2300 |
| }, |
| { |
| "entropy": 0.8104282593354583, |
| "epoch": 3.6959999999999997, |
| "grad_norm": 35.0, |
| "learning_rate": 5.1185185185185195e-06, |
| "loss": 13.0078, |
| "mean_token_accuracy": 0.7940149325877428, |
| "num_tokens": 65696757.0, |
| "step": 2310 |
| }, |
| { |
| "entropy": 0.7942252328619361, |
| "epoch": 3.7119999999999997, |
| "grad_norm": 20.75, |
| "learning_rate": 5.044444444444445e-06, |
| "loss": 12.8079, |
| "mean_token_accuracy": 0.7969783246517181, |
| "num_tokens": 65975610.0, |
| "step": 2320 |
| }, |
| { |
| "entropy": 0.7889417033642531, |
| "epoch": 3.7279999999999998, |
| "grad_norm": 22.0, |
| "learning_rate": 4.970370370370371e-06, |
| "loss": 12.5773, |
| "mean_token_accuracy": 0.7987052239477634, |
| "num_tokens": 66257358.0, |
| "step": 2330 |
| }, |
| { |
| "entropy": 0.7959433317184448, |
| "epoch": 3.7439999999999998, |
| "grad_norm": 19.0, |
| "learning_rate": 4.8962962962962965e-06, |
| "loss": 12.7133, |
| "mean_token_accuracy": 0.7972647827118635, |
| "num_tokens": 66543693.0, |
| "step": 2340 |
| }, |
| { |
| "entropy": 0.7968507738783955, |
| "epoch": 3.76, |
| "grad_norm": 17.125, |
| "learning_rate": 4.822222222222222e-06, |
| "loss": 12.9406, |
| "mean_token_accuracy": 0.7952876027673483, |
| "num_tokens": 66825970.0, |
| "step": 2350 |
| }, |
| { |
| "entropy": 0.8222294202074408, |
| "epoch": 3.776, |
| "grad_norm": 19.625, |
| "learning_rate": 4.748148148148149e-06, |
| "loss": 13.3599, |
| "mean_token_accuracy": 0.7898601289838553, |
| "num_tokens": 67115478.0, |
| "step": 2360 |
| }, |
| { |
| "entropy": 0.7747263446450233, |
| "epoch": 3.792, |
| "grad_norm": 19.75, |
| "learning_rate": 4.674074074074074e-06, |
| "loss": 12.3422, |
| "mean_token_accuracy": 0.8023748748004437, |
| "num_tokens": 67405183.0, |
| "step": 2370 |
| }, |
| { |
| "entropy": 0.8173264717683196, |
| "epoch": 3.808, |
| "grad_norm": 20.75, |
| "learning_rate": 4.600000000000001e-06, |
| "loss": 13.117, |
| "mean_token_accuracy": 0.7941387582570314, |
| "num_tokens": 67688611.0, |
| "step": 2380 |
| }, |
| { |
| "entropy": 0.777202126942575, |
| "epoch": 3.824, |
| "grad_norm": 18.875, |
| "learning_rate": 4.5259259259259265e-06, |
| "loss": 12.5904, |
| "mean_token_accuracy": 0.8000771705061197, |
| "num_tokens": 67974438.0, |
| "step": 2390 |
| }, |
| { |
| "entropy": 0.805173665471375, |
| "epoch": 3.84, |
| "grad_norm": 21.375, |
| "learning_rate": 4.451851851851852e-06, |
| "loss": 12.854, |
| "mean_token_accuracy": 0.7952375136315822, |
| "num_tokens": 68258661.0, |
| "step": 2400 |
| }, |
| { |
| "epoch": 3.84, |
| "eval_biology_entropy": 3.280939736366272, |
| "eval_biology_loss": 3.8342373371124268, |
| "eval_biology_mean_token_accuracy": 0.42561464142799377, |
| "eval_biology_num_tokens": 68258661.0, |
| "eval_biology_runtime": 49.8288, |
| "eval_biology_samples_per_second": 10.034, |
| "eval_biology_steps_per_second": 2.509, |
| "step": 2400 |
| }, |
| { |
| "epoch": 3.84, |
| "eval_chemistry_entropy": 1.8393673224449159, |
| "eval_chemistry_loss": 2.0504181385040283, |
| "eval_chemistry_mean_token_accuracy": 0.6132607061862946, |
| "eval_chemistry_num_tokens": 68258661.0, |
| "eval_chemistry_runtime": 61.5926, |
| "eval_chemistry_samples_per_second": 8.118, |
| "eval_chemistry_steps_per_second": 2.029, |
| "step": 2400 |
| }, |
| { |
| "epoch": 3.84, |
| "eval_math_entropy": 0.8133787260055542, |
| "eval_math_loss": 0.8496425151824951, |
| "eval_math_mean_token_accuracy": 0.7874821076393127, |
| "eval_math_num_tokens": 68258661.0, |
| "eval_math_runtime": 63.1775, |
| "eval_math_samples_per_second": 7.914, |
| "eval_math_steps_per_second": 1.979, |
| "step": 2400 |
| }, |
| { |
| "epoch": 3.84, |
| "eval_physics_entropy": 1.4275367312431335, |
| "eval_physics_loss": 1.5711445808410645, |
| "eval_physics_mean_token_accuracy": 0.6807283799648285, |
| "eval_physics_num_tokens": 68258661.0, |
| "eval_physics_runtime": 71.9466, |
| "eval_physics_samples_per_second": 6.95, |
| "eval_physics_steps_per_second": 1.737, |
| "step": 2400 |
| }, |
| { |
| "entropy": 0.7852308545261621, |
| "epoch": 3.856, |
| "grad_norm": 18.125, |
| "learning_rate": 4.377777777777778e-06, |
| "loss": 12.5377, |
| "mean_token_accuracy": 0.8002926610410214, |
| "num_tokens": 68541830.0, |
| "step": 2410 |
| }, |
| { |
| "entropy": 0.7992656841874123, |
| "epoch": 3.872, |
| "grad_norm": 18.75, |
| "learning_rate": 4.3037037037037035e-06, |
| "loss": 12.8965, |
| "mean_token_accuracy": 0.7948401875793933, |
| "num_tokens": 68832764.0, |
| "step": 2420 |
| }, |
| { |
| "entropy": 0.7982453975826502, |
| "epoch": 3.888, |
| "grad_norm": 19.5, |
| "learning_rate": 4.22962962962963e-06, |
| "loss": 12.8831, |
| "mean_token_accuracy": 0.796014453843236, |
| "num_tokens": 69117158.0, |
| "step": 2430 |
| }, |
| { |
| "entropy": 0.8215716207399965, |
| "epoch": 3.904, |
| "grad_norm": 19.375, |
| "learning_rate": 4.155555555555556e-06, |
| "loss": 13.268, |
| "mean_token_accuracy": 0.7919151060283184, |
| "num_tokens": 69406954.0, |
| "step": 2440 |
| }, |
| { |
| "entropy": 0.784606546163559, |
| "epoch": 3.92, |
| "grad_norm": 20.875, |
| "learning_rate": 4.081481481481482e-06, |
| "loss": 12.5898, |
| "mean_token_accuracy": 0.7998902216553688, |
| "num_tokens": 69682187.0, |
| "step": 2450 |
| }, |
| { |
| "entropy": 0.7975594909861684, |
| "epoch": 3.936, |
| "grad_norm": 20.625, |
| "learning_rate": 4.007407407407408e-06, |
| "loss": 12.8022, |
| "mean_token_accuracy": 0.7962187975645065, |
| "num_tokens": 69977882.0, |
| "step": 2460 |
| }, |
| { |
| "entropy": 0.7960011135786772, |
| "epoch": 3.952, |
| "grad_norm": 17.75, |
| "learning_rate": 3.9333333333333335e-06, |
| "loss": 12.8173, |
| "mean_token_accuracy": 0.7971371866762638, |
| "num_tokens": 70259269.0, |
| "step": 2470 |
| }, |
| { |
| "entropy": 0.8016120189800858, |
| "epoch": 3.968, |
| "grad_norm": 17.875, |
| "learning_rate": 3.85925925925926e-06, |
| "loss": 12.9713, |
| "mean_token_accuracy": 0.7940480273216963, |
| "num_tokens": 70542461.0, |
| "step": 2480 |
| }, |
| { |
| "entropy": 0.7892848145216703, |
| "epoch": 3.984, |
| "grad_norm": 22.375, |
| "learning_rate": 3.7851851851851857e-06, |
| "loss": 12.7155, |
| "mean_token_accuracy": 0.798684723302722, |
| "num_tokens": 70842100.0, |
| "step": 2490 |
| }, |
| { |
| "entropy": 0.799306351877749, |
| "epoch": 4.0, |
| "grad_norm": 19.625, |
| "learning_rate": 3.7111111111111113e-06, |
| "loss": 12.8123, |
| "mean_token_accuracy": 0.7958285242319107, |
| "num_tokens": 71121728.0, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_biology_entropy": 3.2923786249160765, |
| "eval_biology_loss": 3.8357088565826416, |
| "eval_biology_mean_token_accuracy": 0.42535556983947753, |
| "eval_biology_num_tokens": 71121728.0, |
| "eval_biology_runtime": 50.1184, |
| "eval_biology_samples_per_second": 9.976, |
| "eval_biology_steps_per_second": 2.494, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_chemistry_entropy": 1.8444747138023376, |
| "eval_chemistry_loss": 2.050870656967163, |
| "eval_chemistry_mean_token_accuracy": 0.6135786535739899, |
| "eval_chemistry_num_tokens": 71121728.0, |
| "eval_chemistry_runtime": 61.2696, |
| "eval_chemistry_samples_per_second": 8.161, |
| "eval_chemistry_steps_per_second": 2.04, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_math_entropy": 0.8169104042053222, |
| "eval_math_loss": 0.8491711020469666, |
| "eval_math_mean_token_accuracy": 0.7874509425163269, |
| "eval_math_num_tokens": 71121728.0, |
| "eval_math_runtime": 62.7577, |
| "eval_math_samples_per_second": 7.967, |
| "eval_math_steps_per_second": 1.992, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_physics_entropy": 1.4332447266578674, |
| "eval_physics_loss": 1.5686767101287842, |
| "eval_physics_mean_token_accuracy": 0.6809450986385346, |
| "eval_physics_num_tokens": 71121728.0, |
| "eval_physics_runtime": 72.2308, |
| "eval_physics_samples_per_second": 6.922, |
| "eval_physics_steps_per_second": 1.731, |
| "step": 2500 |
| } |
| ], |
| "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": 7.238099893744535e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
