| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.8, |
| "eval_steps": 100, |
| "global_step": 500, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.7213201837614178, |
| "epoch": 0.016, |
| "grad_norm": 127.0, |
| "learning_rate": 6.000000000000001e-07, |
| "loss": 33.4275, |
| "mean_token_accuracy": 0.6860143840312958, |
| "num_tokens": 280941.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.75641950070858, |
| "epoch": 0.032, |
| "grad_norm": 118.0, |
| "learning_rate": 1.2666666666666669e-06, |
| "loss": 33.8085, |
| "mean_token_accuracy": 0.6810433834791183, |
| "num_tokens": 558056.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.7386618180200457, |
| "epoch": 0.048, |
| "grad_norm": 105.5, |
| "learning_rate": 1.9333333333333336e-06, |
| "loss": 33.0823, |
| "mean_token_accuracy": 0.686032148450613, |
| "num_tokens": 836753.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.7653292674571276, |
| "epoch": 0.064, |
| "grad_norm": 106.0, |
| "learning_rate": 2.6e-06, |
| "loss": 33.3421, |
| "mean_token_accuracy": 0.678343191742897, |
| "num_tokens": 1126446.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.7746079994365573, |
| "epoch": 0.08, |
| "grad_norm": 112.0, |
| "learning_rate": 3.266666666666667e-06, |
| "loss": 31.8715, |
| "mean_token_accuracy": 0.6860048722475767, |
| "num_tokens": 1413596.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.7967683909460902, |
| "epoch": 0.096, |
| "grad_norm": 87.5, |
| "learning_rate": 3.9333333333333335e-06, |
| "loss": 30.4245, |
| "mean_token_accuracy": 0.6932173065841198, |
| "num_tokens": 1701193.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.8636196419596672, |
| "epoch": 0.112, |
| "grad_norm": 63.5, |
| "learning_rate": 4.600000000000001e-06, |
| "loss": 28.963, |
| "mean_token_accuracy": 0.6949520386755467, |
| "num_tokens": 1979232.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.986144457757473, |
| "epoch": 0.128, |
| "grad_norm": 45.25, |
| "learning_rate": 5.2666666666666665e-06, |
| "loss": 27.4106, |
| "mean_token_accuracy": 0.7027070496231318, |
| "num_tokens": 2274177.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.0647853959351778, |
| "epoch": 0.144, |
| "grad_norm": 34.5, |
| "learning_rate": 5.933333333333335e-06, |
| "loss": 25.7084, |
| "mean_token_accuracy": 0.7127731442451477, |
| "num_tokens": 2548445.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.129902769997716, |
| "epoch": 0.16, |
| "grad_norm": 29.625, |
| "learning_rate": 6.600000000000001e-06, |
| "loss": 23.6215, |
| "mean_token_accuracy": 0.724789521843195, |
| "num_tokens": 2824418.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 3.7158153285980227, |
| "eval_biology_loss": 4.645914554595947, |
| "eval_biology_mean_token_accuracy": 0.3934867687225342, |
| "eval_biology_num_tokens": 2824418.0, |
| "eval_biology_runtime": 58.324, |
| "eval_biology_samples_per_second": 8.573, |
| "eval_biology_steps_per_second": 2.143, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 2.4825774974823, |
| "eval_chemistry_loss": 2.797250509262085, |
| "eval_chemistry_mean_token_accuracy": 0.5806102266311646, |
| "eval_chemistry_num_tokens": 2824418.0, |
| "eval_chemistry_runtime": 72.1529, |
| "eval_chemistry_samples_per_second": 6.93, |
| "eval_chemistry_steps_per_second": 1.732, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 1.2174424457550048, |
| "eval_math_loss": 1.4766321182250977, |
| "eval_math_mean_token_accuracy": 0.7193697824478149, |
| "eval_math_num_tokens": 2824418.0, |
| "eval_math_runtime": 75.2312, |
| "eval_math_samples_per_second": 6.646, |
| "eval_math_steps_per_second": 1.662, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 2.0236835989952087, |
| "eval_physics_loss": 2.248775005340576, |
| "eval_physics_mean_token_accuracy": 0.6399151833057404, |
| "eval_physics_num_tokens": 2824418.0, |
| "eval_physics_runtime": 85.7293, |
| "eval_physics_samples_per_second": 5.832, |
| "eval_physics_steps_per_second": 1.458, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.1890651997178794, |
| "epoch": 0.176, |
| "grad_norm": 24.5, |
| "learning_rate": 7.266666666666668e-06, |
| "loss": 22.5161, |
| "mean_token_accuracy": 0.7291781000792981, |
| "num_tokens": 3110313.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.2231741398572922, |
| "epoch": 0.192, |
| "grad_norm": 19.375, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 22.3058, |
| "mean_token_accuracy": 0.7288031376898289, |
| "num_tokens": 3394170.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.1990749202668667, |
| "epoch": 0.208, |
| "grad_norm": 17.125, |
| "learning_rate": 8.6e-06, |
| "loss": 21.6321, |
| "mean_token_accuracy": 0.7331194914877415, |
| "num_tokens": 3673600.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.1935963843017816, |
| "epoch": 0.224, |
| "grad_norm": 18.5, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 20.9973, |
| "mean_token_accuracy": 0.7374927394092083, |
| "num_tokens": 3953732.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.1516071777790784, |
| "epoch": 0.24, |
| "grad_norm": 15.6875, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 20.0443, |
| "mean_token_accuracy": 0.7471608892083168, |
| "num_tokens": 4243655.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.1274361703544855, |
| "epoch": 0.256, |
| "grad_norm": 17.125, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 19.5101, |
| "mean_token_accuracy": 0.7516594093292952, |
| "num_tokens": 4531471.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.140014799311757, |
| "epoch": 0.272, |
| "grad_norm": 14.0, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 19.5844, |
| "mean_token_accuracy": 0.7498023040592671, |
| "num_tokens": 4810284.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.1203772608190774, |
| "epoch": 0.288, |
| "grad_norm": 12.5625, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 19.1823, |
| "mean_token_accuracy": 0.753832882642746, |
| "num_tokens": 5095104.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.129030792787671, |
| "epoch": 0.304, |
| "grad_norm": 12.9375, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 19.1862, |
| "mean_token_accuracy": 0.7524638257920742, |
| "num_tokens": 5383732.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.090552070736885, |
| "epoch": 0.32, |
| "grad_norm": 15.0625, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 18.5807, |
| "mean_token_accuracy": 0.7593985345214606, |
| "num_tokens": 5676334.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 3.458996726989746, |
| "eval_biology_loss": 4.269806385040283, |
| "eval_biology_mean_token_accuracy": 0.4222190487384796, |
| "eval_biology_num_tokens": 5676334.0, |
| "eval_biology_runtime": 58.3189, |
| "eval_biology_samples_per_second": 8.574, |
| "eval_biology_steps_per_second": 2.143, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 2.180737380027771, |
| "eval_chemistry_loss": 2.439823865890503, |
| "eval_chemistry_mean_token_accuracy": 0.6122945117950439, |
| "eval_chemistry_num_tokens": 5676334.0, |
| "eval_chemistry_runtime": 72.4241, |
| "eval_chemistry_samples_per_second": 6.904, |
| "eval_chemistry_steps_per_second": 1.726, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 1.1109818544387817, |
| "eval_math_loss": 1.1917932033538818, |
| "eval_math_mean_token_accuracy": 0.752472719669342, |
| "eval_math_num_tokens": 5676334.0, |
| "eval_math_runtime": 75.8098, |
| "eval_math_samples_per_second": 6.595, |
| "eval_math_steps_per_second": 1.649, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 1.7737830457687378, |
| "eval_physics_loss": 1.935411810874939, |
| "eval_physics_mean_token_accuracy": 0.670736522436142, |
| "eval_physics_num_tokens": 5676334.0, |
| "eval_physics_runtime": 86.0358, |
| "eval_physics_samples_per_second": 5.812, |
| "eval_physics_steps_per_second": 1.453, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.1083258152008058, |
| "epoch": 0.336, |
| "grad_norm": 14.875, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 18.9806, |
| "mean_token_accuracy": 0.7541178163141012, |
| "num_tokens": 5958480.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.0838440161198377, |
| "epoch": 0.352, |
| "grad_norm": 13.625, |
| "learning_rate": 1.46e-05, |
| "loss": 18.2469, |
| "mean_token_accuracy": 0.7600458469241858, |
| "num_tokens": 6242161.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 1.0777738582342864, |
| "epoch": 0.368, |
| "grad_norm": 14.6875, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 18.3345, |
| "mean_token_accuracy": 0.759134492278099, |
| "num_tokens": 6523679.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.0698408473283052, |
| "epoch": 0.384, |
| "grad_norm": 14.0, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 17.9275, |
| "mean_token_accuracy": 0.7646602354943752, |
| "num_tokens": 6810978.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.0921947304159403, |
| "epoch": 0.4, |
| "grad_norm": 13.8125, |
| "learning_rate": 1.66e-05, |
| "loss": 18.5932, |
| "mean_token_accuracy": 0.757987079769373, |
| "num_tokens": 7096903.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 1.0657698467373848, |
| "epoch": 0.416, |
| "grad_norm": 12.9375, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 17.8874, |
| "mean_token_accuracy": 0.7630033176392317, |
| "num_tokens": 7377181.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 1.0416713248938323, |
| "epoch": 0.432, |
| "grad_norm": 15.1875, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 17.5118, |
| "mean_token_accuracy": 0.7672814972698688, |
| "num_tokens": 7650523.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 1.0186962354928255, |
| "epoch": 0.448, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.86e-05, |
| "loss": 17.3118, |
| "mean_token_accuracy": 0.7682009607553482, |
| "num_tokens": 7936788.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 1.036542361229658, |
| "epoch": 0.464, |
| "grad_norm": 15.8125, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 17.3748, |
| "mean_token_accuracy": 0.7694676879793405, |
| "num_tokens": 8223147.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 1.0569938763976097, |
| "epoch": 0.48, |
| "grad_norm": 14.5, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 17.6393, |
| "mean_token_accuracy": 0.7655092909932136, |
| "num_tokens": 8506030.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 3.567513717651367, |
| "eval_biology_loss": 4.103325366973877, |
| "eval_biology_mean_token_accuracy": 0.43289893221855164, |
| "eval_biology_num_tokens": 8506030.0, |
| "eval_biology_runtime": 58.2179, |
| "eval_biology_samples_per_second": 8.588, |
| "eval_biology_steps_per_second": 2.147, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 2.159853268146515, |
| "eval_chemistry_loss": 2.3111026287078857, |
| "eval_chemistry_mean_token_accuracy": 0.6211932191848755, |
| "eval_chemistry_num_tokens": 8506030.0, |
| "eval_chemistry_runtime": 72.4538, |
| "eval_chemistry_samples_per_second": 6.901, |
| "eval_chemistry_steps_per_second": 1.725, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 1.055273208618164, |
| "eval_math_loss": 1.103938341140747, |
| "eval_math_mean_token_accuracy": 0.7635622372627259, |
| "eval_math_num_tokens": 8506030.0, |
| "eval_math_runtime": 75.7526, |
| "eval_math_samples_per_second": 6.6, |
| "eval_math_steps_per_second": 1.65, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 1.7374450120925904, |
| "eval_physics_loss": 1.8264347314834595, |
| "eval_physics_mean_token_accuracy": 0.6800054814815522, |
| "eval_physics_num_tokens": 8506030.0, |
| "eval_physics_runtime": 85.6984, |
| "eval_physics_samples_per_second": 5.834, |
| "eval_physics_steps_per_second": 1.459, |
| "step": 300 |
| }, |
| { |
| "entropy": 1.0048395186662673, |
| "epoch": 0.496, |
| "grad_norm": 20.75, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 17.0439, |
| "mean_token_accuracy": 0.7706491403281689, |
| "num_tokens": 8788726.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 1.0206645488739015, |
| "epoch": 0.512, |
| "grad_norm": 14.125, |
| "learning_rate": 1.985925925925926e-05, |
| "loss": 17.0396, |
| "mean_token_accuracy": 0.7719013765454292, |
| "num_tokens": 9078039.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.9999576449394226, |
| "epoch": 0.528, |
| "grad_norm": 15.0625, |
| "learning_rate": 1.9785185185185187e-05, |
| "loss": 16.7994, |
| "mean_token_accuracy": 0.7749712273478508, |
| "num_tokens": 9373860.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 1.0083586767315864, |
| "epoch": 0.544, |
| "grad_norm": 15.0625, |
| "learning_rate": 1.971111111111111e-05, |
| "loss": 16.9209, |
| "mean_token_accuracy": 0.7731953993439674, |
| "num_tokens": 9660940.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 1.0532048616558314, |
| "epoch": 0.56, |
| "grad_norm": 18.75, |
| "learning_rate": 1.963703703703704e-05, |
| "loss": 17.638, |
| "mean_token_accuracy": 0.7637613207101822, |
| "num_tokens": 9932302.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 1.0087582137435676, |
| "epoch": 0.576, |
| "grad_norm": 17.75, |
| "learning_rate": 1.9562962962962964e-05, |
| "loss": 17.0002, |
| "mean_token_accuracy": 0.7717114143073559, |
| "num_tokens": 10215462.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 1.0042562756687403, |
| "epoch": 0.592, |
| "grad_norm": 15.0, |
| "learning_rate": 1.948888888888889e-05, |
| "loss": 16.7266, |
| "mean_token_accuracy": 0.7740717619657517, |
| "num_tokens": 10504396.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.9980081049725413, |
| "epoch": 0.608, |
| "grad_norm": 16.125, |
| "learning_rate": 1.9414814814814817e-05, |
| "loss": 16.8411, |
| "mean_token_accuracy": 0.7726139184087515, |
| "num_tokens": 10793126.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.9865465115755796, |
| "epoch": 0.624, |
| "grad_norm": 14.5, |
| "learning_rate": 1.9340740740740743e-05, |
| "loss": 16.5077, |
| "mean_token_accuracy": 0.7759908959269524, |
| "num_tokens": 11081768.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.9753864344209433, |
| "epoch": 0.64, |
| "grad_norm": 13.0625, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 16.3328, |
| "mean_token_accuracy": 0.777353797480464, |
| "num_tokens": 11370320.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 3.4534600849151613, |
| "eval_biology_loss": 4.02693510055542, |
| "eval_biology_mean_token_accuracy": 0.4380790627002716, |
| "eval_biology_num_tokens": 11370320.0, |
| "eval_biology_runtime": 58.2953, |
| "eval_biology_samples_per_second": 8.577, |
| "eval_biology_steps_per_second": 2.144, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 2.068386361122131, |
| "eval_chemistry_loss": 2.252868175506592, |
| "eval_chemistry_mean_token_accuracy": 0.6284125361442566, |
| "eval_chemistry_num_tokens": 11370320.0, |
| "eval_chemistry_runtime": 72.4168, |
| "eval_chemistry_samples_per_second": 6.904, |
| "eval_chemistry_steps_per_second": 1.726, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 1.008522126674652, |
| "eval_math_loss": 1.0611059665679932, |
| "eval_math_mean_token_accuracy": 0.7697986388206481, |
| "eval_math_num_tokens": 11370320.0, |
| "eval_math_runtime": 75.8656, |
| "eval_math_samples_per_second": 6.591, |
| "eval_math_steps_per_second": 1.648, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 1.6607081370353698, |
| "eval_physics_loss": 1.7777302265167236, |
| "eval_physics_mean_token_accuracy": 0.685634191274643, |
| "eval_physics_num_tokens": 11370320.0, |
| "eval_physics_runtime": 86.0729, |
| "eval_physics_samples_per_second": 5.809, |
| "eval_physics_steps_per_second": 1.452, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.9546368110924959, |
| "epoch": 0.656, |
| "grad_norm": 13.3125, |
| "learning_rate": 1.9192592592592593e-05, |
| "loss": 16.1585, |
| "mean_token_accuracy": 0.7791520744562149, |
| "num_tokens": 11657835.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 1.0121946074068546, |
| "epoch": 0.672, |
| "grad_norm": 14.9375, |
| "learning_rate": 1.911851851851852e-05, |
| "loss": 16.7638, |
| "mean_token_accuracy": 0.7723276257514954, |
| "num_tokens": 11949262.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.9564321564510465, |
| "epoch": 0.688, |
| "grad_norm": 12.875, |
| "learning_rate": 1.9044444444444446e-05, |
| "loss": 16.2106, |
| "mean_token_accuracy": 0.7782567508518696, |
| "num_tokens": 12227640.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.9711398147046566, |
| "epoch": 0.704, |
| "grad_norm": 12.875, |
| "learning_rate": 1.8970370370370372e-05, |
| "loss": 16.2595, |
| "mean_token_accuracy": 0.7794518828392029, |
| "num_tokens": 12516641.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.990043681114912, |
| "epoch": 0.72, |
| "grad_norm": 16.0, |
| "learning_rate": 1.8896296296296295e-05, |
| "loss": 16.6764, |
| "mean_token_accuracy": 0.7741847857832909, |
| "num_tokens": 12793343.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.9794478964060545, |
| "epoch": 0.736, |
| "grad_norm": 13.5625, |
| "learning_rate": 1.8822222222222225e-05, |
| "loss": 16.3042, |
| "mean_token_accuracy": 0.7780686043202877, |
| "num_tokens": 13077981.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.9613467682152986, |
| "epoch": 0.752, |
| "grad_norm": 14.6875, |
| "learning_rate": 1.874814814814815e-05, |
| "loss": 16.0345, |
| "mean_token_accuracy": 0.7808727387338876, |
| "num_tokens": 13358957.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.9237693391740323, |
| "epoch": 0.768, |
| "grad_norm": 12.0, |
| "learning_rate": 1.8674074074074075e-05, |
| "loss": 15.6444, |
| "mean_token_accuracy": 0.7876643743366003, |
| "num_tokens": 13653412.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 1.0019762709736824, |
| "epoch": 0.784, |
| "grad_norm": 15.6875, |
| "learning_rate": 1.86e-05, |
| "loss": 16.7034, |
| "mean_token_accuracy": 0.7748573537915945, |
| "num_tokens": 13940856.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.9975844457745552, |
| "epoch": 0.8, |
| "grad_norm": 17.0, |
| "learning_rate": 1.8525925925925928e-05, |
| "loss": 16.7128, |
| "mean_token_accuracy": 0.77371263243258, |
| "num_tokens": 14230754.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 3.498454050064087, |
| "eval_biology_loss": 3.949568271636963, |
| "eval_biology_mean_token_accuracy": 0.44270548820495603, |
| "eval_biology_num_tokens": 14230754.0, |
| "eval_biology_runtime": 58.3602, |
| "eval_biology_samples_per_second": 8.567, |
| "eval_biology_steps_per_second": 2.142, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 2.073633497238159, |
| "eval_chemistry_loss": 2.1966874599456787, |
| "eval_chemistry_mean_token_accuracy": 0.6324296245574951, |
| "eval_chemistry_num_tokens": 14230754.0, |
| "eval_chemistry_runtime": 71.3745, |
| "eval_chemistry_samples_per_second": 7.005, |
| "eval_chemistry_steps_per_second": 1.751, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 0.9948915004730224, |
| "eval_math_loss": 1.0365564823150635, |
| "eval_math_mean_token_accuracy": 0.7735790548324585, |
| "eval_math_num_tokens": 14230754.0, |
| "eval_math_runtime": 75.2012, |
| "eval_math_samples_per_second": 6.649, |
| "eval_math_steps_per_second": 1.662, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 1.6648971457481385, |
| "eval_physics_loss": 1.736081838607788, |
| "eval_physics_mean_token_accuracy": 0.6900339169502259, |
| "eval_physics_num_tokens": 14230754.0, |
| "eval_physics_runtime": 84.9467, |
| "eval_physics_samples_per_second": 5.886, |
| "eval_physics_steps_per_second": 1.472, |
| "step": 500 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 3000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.448451560178856e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
