| { |
| "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": 0.7081545192748309, |
| "epoch": 0.016, |
| "grad_norm": 119.5, |
| "learning_rate": 6.000000000000001e-07, |
| "loss": 30.4211, |
| "mean_token_accuracy": 0.7030085142701864, |
| "num_tokens": 338352.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.7043411549180746, |
| "epoch": 0.032, |
| "grad_norm": 111.5, |
| "learning_rate": 1.2666666666666669e-06, |
| "loss": 30.1214, |
| "mean_token_accuracy": 0.7080705553293228, |
| "num_tokens": 671193.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.7044402219355106, |
| "epoch": 0.048, |
| "grad_norm": 106.0, |
| "learning_rate": 1.9333333333333336e-06, |
| "loss": 29.7341, |
| "mean_token_accuracy": 0.7064134150743484, |
| "num_tokens": 1001008.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.733870861120522, |
| "epoch": 0.064, |
| "grad_norm": 96.0, |
| "learning_rate": 2.6e-06, |
| "loss": 30.1494, |
| "mean_token_accuracy": 0.7005880128592252, |
| "num_tokens": 1331208.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.7398215295746923, |
| "epoch": 0.08, |
| "grad_norm": 90.5, |
| "learning_rate": 3.266666666666667e-06, |
| "loss": 28.7202, |
| "mean_token_accuracy": 0.7051243670284748, |
| "num_tokens": 1667532.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.738157893717289, |
| "epoch": 0.096, |
| "grad_norm": 76.5, |
| "learning_rate": 3.9333333333333335e-06, |
| "loss": 27.1633, |
| "mean_token_accuracy": 0.71303277797997, |
| "num_tokens": 2007176.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.8131938450038433, |
| "epoch": 0.112, |
| "grad_norm": 57.5, |
| "learning_rate": 4.600000000000001e-06, |
| "loss": 26.4428, |
| "mean_token_accuracy": 0.7120848346501589, |
| "num_tokens": 2342259.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.9618701986968518, |
| "epoch": 0.128, |
| "grad_norm": 43.75, |
| "learning_rate": 5.2666666666666665e-06, |
| "loss": 25.9908, |
| "mean_token_accuracy": 0.7077770136296749, |
| "num_tokens": 2663987.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.0031472396105527, |
| "epoch": 0.144, |
| "grad_norm": 30.125, |
| "learning_rate": 5.933333333333335e-06, |
| "loss": 23.3524, |
| "mean_token_accuracy": 0.7268724206835031, |
| "num_tokens": 2997824.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.0786659345030785, |
| "epoch": 0.16, |
| "grad_norm": 20.75, |
| "learning_rate": 6.600000000000001e-06, |
| "loss": 21.8746, |
| "mean_token_accuracy": 0.7358953565359115, |
| "num_tokens": 3330597.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 3.260982286453247, |
| "eval_biology_loss": 3.7093346118927, |
| "eval_biology_mean_token_accuracy": 0.46263515305519104, |
| "eval_biology_num_tokens": 3330597.0, |
| "eval_biology_runtime": 57.0058, |
| "eval_biology_samples_per_second": 8.771, |
| "eval_biology_steps_per_second": 2.193, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 1.6407185397148132, |
| "eval_chemistry_loss": 1.7927597761154175, |
| "eval_chemistry_mean_token_accuracy": 0.677183347940445, |
| "eval_chemistry_num_tokens": 3330597.0, |
| "eval_chemistry_runtime": 71.6705, |
| "eval_chemistry_samples_per_second": 6.976, |
| "eval_chemistry_steps_per_second": 1.744, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 0.9620108480453491, |
| "eval_math_loss": 1.4342334270477295, |
| "eval_math_mean_token_accuracy": 0.7288489332199096, |
| "eval_math_num_tokens": 3330597.0, |
| "eval_math_runtime": 74.2141, |
| "eval_math_samples_per_second": 6.737, |
| "eval_math_steps_per_second": 1.684, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 1.1280243759155273, |
| "eval_physics_loss": 1.3446351289749146, |
| "eval_physics_mean_token_accuracy": 0.7332401766777038, |
| "eval_physics_num_tokens": 3330597.0, |
| "eval_physics_runtime": 84.9827, |
| "eval_physics_samples_per_second": 5.884, |
| "eval_physics_steps_per_second": 1.471, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.1533773414790631, |
| "epoch": 0.176, |
| "grad_norm": 19.75, |
| "learning_rate": 7.266666666666668e-06, |
| "loss": 21.3285, |
| "mean_token_accuracy": 0.7356826655566693, |
| "num_tokens": 3658264.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.112282781302929, |
| "epoch": 0.192, |
| "grad_norm": 12.625, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 20.4084, |
| "mean_token_accuracy": 0.7426700782030821, |
| "num_tokens": 3995568.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.086438114568591, |
| "epoch": 0.208, |
| "grad_norm": 13.875, |
| "learning_rate": 8.6e-06, |
| "loss": 19.3962, |
| "mean_token_accuracy": 0.7525120593607426, |
| "num_tokens": 4321436.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.089601282030344, |
| "epoch": 0.224, |
| "grad_norm": 11.8125, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 19.0472, |
| "mean_token_accuracy": 0.7530231148004531, |
| "num_tokens": 4648491.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.0804125092923642, |
| "epoch": 0.24, |
| "grad_norm": 11.8125, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 18.7739, |
| "mean_token_accuracy": 0.7554640270769596, |
| "num_tokens": 4986175.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.0836318269371987, |
| "epoch": 0.256, |
| "grad_norm": 11.9375, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 18.4704, |
| "mean_token_accuracy": 0.7584241010248661, |
| "num_tokens": 5329320.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.0882270745933056, |
| "epoch": 0.272, |
| "grad_norm": 13.375, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 18.6237, |
| "mean_token_accuracy": 0.754904105886817, |
| "num_tokens": 5658796.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.076054659858346, |
| "epoch": 0.288, |
| "grad_norm": 11.0625, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 18.4139, |
| "mean_token_accuracy": 0.7581049785017967, |
| "num_tokens": 5980201.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.072243456915021, |
| "epoch": 0.304, |
| "grad_norm": 15.5, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 18.1532, |
| "mean_token_accuracy": 0.7589787419885397, |
| "num_tokens": 6298858.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.0509168311953545, |
| "epoch": 0.32, |
| "grad_norm": 12.0, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 17.8979, |
| "mean_token_accuracy": 0.7612847778946161, |
| "num_tokens": 6622798.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 2.7866374349594114, |
| "eval_biology_loss": 3.2049155235290527, |
| "eval_biology_mean_token_accuracy": 0.502942953824997, |
| "eval_biology_num_tokens": 6622798.0, |
| "eval_biology_runtime": 56.7587, |
| "eval_biology_samples_per_second": 8.809, |
| "eval_biology_steps_per_second": 2.202, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 1.41602649641037, |
| "eval_chemistry_loss": 1.4916778802871704, |
| "eval_chemistry_mean_token_accuracy": 0.7079435839653015, |
| "eval_chemistry_num_tokens": 6622798.0, |
| "eval_chemistry_runtime": 71.0228, |
| "eval_chemistry_samples_per_second": 7.04, |
| "eval_chemistry_steps_per_second": 1.76, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 0.9821910600662231, |
| "eval_math_loss": 1.256189227104187, |
| "eval_math_mean_token_accuracy": 0.7425830450057983, |
| "eval_math_num_tokens": 6622798.0, |
| "eval_math_runtime": 75.4477, |
| "eval_math_samples_per_second": 6.627, |
| "eval_math_steps_per_second": 1.657, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 1.0466554942131043, |
| "eval_physics_loss": 1.1129223108291626, |
| "eval_physics_mean_token_accuracy": 0.7617223796844482, |
| "eval_physics_num_tokens": 6622798.0, |
| "eval_physics_runtime": 85.3386, |
| "eval_physics_samples_per_second": 5.859, |
| "eval_physics_steps_per_second": 1.465, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.0462193094193935, |
| "epoch": 0.336, |
| "grad_norm": 11.875, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 17.7296, |
| "mean_token_accuracy": 0.7623197592794895, |
| "num_tokens": 6953611.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.0164767142385245, |
| "epoch": 0.352, |
| "grad_norm": 11.25, |
| "learning_rate": 1.46e-05, |
| "loss": 17.1525, |
| "mean_token_accuracy": 0.7695928748697043, |
| "num_tokens": 7290378.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.9987690806388855, |
| "epoch": 0.368, |
| "grad_norm": 11.4375, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 16.9228, |
| "mean_token_accuracy": 0.7714470315724611, |
| "num_tokens": 7621082.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.0506686847656965, |
| "epoch": 0.384, |
| "grad_norm": 11.3125, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 17.6416, |
| "mean_token_accuracy": 0.7599896423518657, |
| "num_tokens": 7955570.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.0135670874267817, |
| "epoch": 0.4, |
| "grad_norm": 14.5, |
| "learning_rate": 1.66e-05, |
| "loss": 16.9698, |
| "mean_token_accuracy": 0.7687441952526569, |
| "num_tokens": 8291049.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.9876859273761511, |
| "epoch": 0.416, |
| "grad_norm": 13.75, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 16.6909, |
| "mean_token_accuracy": 0.7725014701485634, |
| "num_tokens": 8627310.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 1.0216444052755833, |
| "epoch": 0.432, |
| "grad_norm": 12.125, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 17.011, |
| "mean_token_accuracy": 0.7689431305974722, |
| "num_tokens": 8958371.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.9649551097303629, |
| "epoch": 0.448, |
| "grad_norm": 12.1875, |
| "learning_rate": 1.86e-05, |
| "loss": 16.2252, |
| "mean_token_accuracy": 0.7780237648636102, |
| "num_tokens": 9294388.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.9653072291985154, |
| "epoch": 0.464, |
| "grad_norm": 12.5, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 16.1144, |
| "mean_token_accuracy": 0.7772431872785092, |
| "num_tokens": 9620366.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.9695413928478956, |
| "epoch": 0.48, |
| "grad_norm": 11.8125, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 16.4447, |
| "mean_token_accuracy": 0.7733453687280416, |
| "num_tokens": 9955431.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 2.7462952222824097, |
| "eval_biology_loss": 3.060053825378418, |
| "eval_biology_mean_token_accuracy": 0.5114872539043427, |
| "eval_biology_num_tokens": 9955431.0, |
| "eval_biology_runtime": 56.9805, |
| "eval_biology_samples_per_second": 8.775, |
| "eval_biology_steps_per_second": 2.194, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 1.3710836601257324, |
| "eval_chemistry_loss": 1.4066872596740723, |
| "eval_chemistry_mean_token_accuracy": 0.7168692064285278, |
| "eval_chemistry_num_tokens": 9955431.0, |
| "eval_chemistry_runtime": 71.3717, |
| "eval_chemistry_samples_per_second": 7.006, |
| "eval_chemistry_steps_per_second": 1.751, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 0.9784291224479675, |
| "eval_math_loss": 1.2082456350326538, |
| "eval_math_mean_token_accuracy": 0.7478420066833497, |
| "eval_math_num_tokens": 9955431.0, |
| "eval_math_runtime": 74.3706, |
| "eval_math_samples_per_second": 6.723, |
| "eval_math_steps_per_second": 1.681, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 1.007008915424347, |
| "eval_physics_loss": 1.0401662588119507, |
| "eval_physics_mean_token_accuracy": 0.7707204632759094, |
| "eval_physics_num_tokens": 9955431.0, |
| "eval_physics_runtime": 84.6939, |
| "eval_physics_samples_per_second": 5.904, |
| "eval_physics_steps_per_second": 1.476, |
| "step": 300 |
| }, |
| { |
| "entropy": 1.007104156538844, |
| "epoch": 0.496, |
| "grad_norm": 10.5, |
| "learning_rate": 1.9933333333333334e-05, |
| "loss": 16.6639, |
| "mean_token_accuracy": 0.771253039687872, |
| "num_tokens": 10289883.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.9723599992692471, |
| "epoch": 0.512, |
| "grad_norm": 12.4375, |
| "learning_rate": 1.985925925925926e-05, |
| "loss": 16.4387, |
| "mean_token_accuracy": 0.7748051699250936, |
| "num_tokens": 10619468.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.9979474730789661, |
| "epoch": 0.528, |
| "grad_norm": 10.8125, |
| "learning_rate": 1.9785185185185187e-05, |
| "loss": 16.6398, |
| "mean_token_accuracy": 0.7712166555225849, |
| "num_tokens": 10944904.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.9800913408398628, |
| "epoch": 0.544, |
| "grad_norm": 11.625, |
| "learning_rate": 1.971111111111111e-05, |
| "loss": 16.374, |
| "mean_token_accuracy": 0.7737775973975658, |
| "num_tokens": 11266409.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.9599094491451978, |
| "epoch": 0.56, |
| "grad_norm": 10.625, |
| "learning_rate": 1.963703703703704e-05, |
| "loss": 16.0194, |
| "mean_token_accuracy": 0.778597004711628, |
| "num_tokens": 11605544.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.948445113003254, |
| "epoch": 0.576, |
| "grad_norm": 11.75, |
| "learning_rate": 1.9562962962962964e-05, |
| "loss": 15.9791, |
| "mean_token_accuracy": 0.7805619373917579, |
| "num_tokens": 11931715.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.9618662863969802, |
| "epoch": 0.592, |
| "grad_norm": 13.0625, |
| "learning_rate": 1.948888888888889e-05, |
| "loss": 16.0713, |
| "mean_token_accuracy": 0.776887471601367, |
| "num_tokens": 12257243.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.9480277199298144, |
| "epoch": 0.608, |
| "grad_norm": 10.8125, |
| "learning_rate": 1.9414814814814817e-05, |
| "loss": 15.8656, |
| "mean_token_accuracy": 0.7797395702451467, |
| "num_tokens": 12583154.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.9716805059462785, |
| "epoch": 0.624, |
| "grad_norm": 10.625, |
| "learning_rate": 1.9340740740740743e-05, |
| "loss": 16.1886, |
| "mean_token_accuracy": 0.7762768242508173, |
| "num_tokens": 12905392.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.9691856369376183, |
| "epoch": 0.64, |
| "grad_norm": 11.8125, |
| "learning_rate": 1.926666666666667e-05, |
| "loss": 16.2314, |
| "mean_token_accuracy": 0.7761619180440903, |
| "num_tokens": 13232198.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 2.67273916721344, |
| "eval_biology_loss": 2.96358585357666, |
| "eval_biology_mean_token_accuracy": 0.5216054661273957, |
| "eval_biology_num_tokens": 13232198.0, |
| "eval_biology_runtime": 57.8851, |
| "eval_biology_samples_per_second": 8.638, |
| "eval_biology_steps_per_second": 2.159, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 1.3173069462776184, |
| "eval_chemistry_loss": 1.36112380027771, |
| "eval_chemistry_mean_token_accuracy": 0.7232093935012818, |
| "eval_chemistry_num_tokens": 13232198.0, |
| "eval_chemistry_runtime": 71.8687, |
| "eval_chemistry_samples_per_second": 6.957, |
| "eval_chemistry_steps_per_second": 1.739, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 0.9528153438568115, |
| "eval_math_loss": 1.1841319799423218, |
| "eval_math_mean_token_accuracy": 0.7507999067306519, |
| "eval_math_num_tokens": 13232198.0, |
| "eval_math_runtime": 74.8756, |
| "eval_math_samples_per_second": 6.678, |
| "eval_math_steps_per_second": 1.669, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 0.9614342203140259, |
| "eval_physics_loss": 1.0016775131225586, |
| "eval_physics_mean_token_accuracy": 0.7765112986564636, |
| "eval_physics_num_tokens": 13232198.0, |
| "eval_physics_runtime": 85.1175, |
| "eval_physics_samples_per_second": 5.874, |
| "eval_physics_steps_per_second": 1.469, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.9530647125095129, |
| "epoch": 0.656, |
| "grad_norm": 11.375, |
| "learning_rate": 1.9192592592592593e-05, |
| "loss": 15.9429, |
| "mean_token_accuracy": 0.7780249908566474, |
| "num_tokens": 13575902.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.9399624351412058, |
| "epoch": 0.672, |
| "grad_norm": 12.0625, |
| "learning_rate": 1.911851851851852e-05, |
| "loss": 15.7284, |
| "mean_token_accuracy": 0.7817466016858816, |
| "num_tokens": 13895997.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.9540142439305782, |
| "epoch": 0.688, |
| "grad_norm": 10.6875, |
| "learning_rate": 1.9044444444444446e-05, |
| "loss": 15.7763, |
| "mean_token_accuracy": 0.7817316662520171, |
| "num_tokens": 14234888.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.9538515221327544, |
| "epoch": 0.704, |
| "grad_norm": 11.625, |
| "learning_rate": 1.8970370370370372e-05, |
| "loss": 16.0838, |
| "mean_token_accuracy": 0.775813952833414, |
| "num_tokens": 14567908.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.9464580094441771, |
| "epoch": 0.72, |
| "grad_norm": 12.375, |
| "learning_rate": 1.8896296296296295e-05, |
| "loss": 15.704, |
| "mean_token_accuracy": 0.7821726374328136, |
| "num_tokens": 14882927.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.9363573594018817, |
| "epoch": 0.736, |
| "grad_norm": 13.0625, |
| "learning_rate": 1.8822222222222225e-05, |
| "loss": 15.7075, |
| "mean_token_accuracy": 0.7809599131345749, |
| "num_tokens": 15217342.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.9193624388426542, |
| "epoch": 0.752, |
| "grad_norm": 10.4375, |
| "learning_rate": 1.874814814814815e-05, |
| "loss": 15.3677, |
| "mean_token_accuracy": 0.7846659943461418, |
| "num_tokens": 15547710.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.9613647162914276, |
| "epoch": 0.768, |
| "grad_norm": 11.1875, |
| "learning_rate": 1.8674074074074075e-05, |
| "loss": 16.0934, |
| "mean_token_accuracy": 0.7759233452379704, |
| "num_tokens": 15877177.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.9139260027557612, |
| "epoch": 0.784, |
| "grad_norm": 10.1875, |
| "learning_rate": 1.86e-05, |
| "loss": 15.3216, |
| "mean_token_accuracy": 0.786144433543086, |
| "num_tokens": 16219640.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.9349881060421467, |
| "epoch": 0.8, |
| "grad_norm": 12.0, |
| "learning_rate": 1.8525925925925928e-05, |
| "loss": 15.6534, |
| "mean_token_accuracy": 0.7819662269204855, |
| "num_tokens": 16548261.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 2.617759398460388, |
| "eval_biology_loss": 2.9445557594299316, |
| "eval_biology_mean_token_accuracy": 0.5238966343402862, |
| "eval_biology_num_tokens": 16548261.0, |
| "eval_biology_runtime": 57.6905, |
| "eval_biology_samples_per_second": 8.667, |
| "eval_biology_steps_per_second": 2.167, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 1.2937732858657838, |
| "eval_chemistry_loss": 1.3414490222930908, |
| "eval_chemistry_mean_token_accuracy": 0.7248484263420105, |
| "eval_chemistry_num_tokens": 16548261.0, |
| "eval_chemistry_runtime": 71.1659, |
| "eval_chemistry_samples_per_second": 7.026, |
| "eval_chemistry_steps_per_second": 1.756, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 0.93889684009552, |
| "eval_math_loss": 1.17385995388031, |
| "eval_math_mean_token_accuracy": 0.7522515163421631, |
| "eval_math_num_tokens": 16548261.0, |
| "eval_math_runtime": 74.7142, |
| "eval_math_samples_per_second": 6.692, |
| "eval_math_steps_per_second": 1.673, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 0.9420124840736389, |
| "eval_physics_loss": 0.9809063673019409, |
| "eval_physics_mean_token_accuracy": 0.7797514395713806, |
| "eval_physics_num_tokens": 16548261.0, |
| "eval_physics_runtime": 84.5204, |
| "eval_physics_samples_per_second": 5.916, |
| "eval_physics_steps_per_second": 1.479, |
| "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": 1.652179086470234e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
