| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.2, |
| "eval_steps": 100, |
| "global_step": 2000, |
| "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 |
| }, |
| { |
| "entropy": 0.9288215478882194, |
| "epoch": 0.816, |
| "grad_norm": 11.3125, |
| "learning_rate": 1.8451851851851855e-05, |
| "loss": 15.5725, |
| "mean_token_accuracy": 0.7813547737896442, |
| "num_tokens": 16870533.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.9446330968290567, |
| "epoch": 0.832, |
| "grad_norm": 15.875, |
| "learning_rate": 1.8377777777777778e-05, |
| "loss": 15.7952, |
| "mean_token_accuracy": 0.7800830528140068, |
| "num_tokens": 17196371.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.9318377874791622, |
| "epoch": 0.848, |
| "grad_norm": 10.625, |
| "learning_rate": 1.8303703703703704e-05, |
| "loss": 15.4484, |
| "mean_token_accuracy": 0.7836398217827082, |
| "num_tokens": 17527066.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.9298348639160394, |
| "epoch": 0.864, |
| "grad_norm": 10.625, |
| "learning_rate": 1.822962962962963e-05, |
| "loss": 15.6026, |
| "mean_token_accuracy": 0.7814645994454622, |
| "num_tokens": 17859605.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.9549105998128653, |
| "epoch": 0.88, |
| "grad_norm": 12.625, |
| "learning_rate": 1.8155555555555557e-05, |
| "loss": 15.7997, |
| "mean_token_accuracy": 0.7805559232831001, |
| "num_tokens": 18187315.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.8820257507264614, |
| "epoch": 0.896, |
| "grad_norm": 10.5625, |
| "learning_rate": 1.8081481481481484e-05, |
| "loss": 14.7783, |
| "mean_token_accuracy": 0.7907331828027964, |
| "num_tokens": 18536790.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.9362711973488331, |
| "epoch": 0.912, |
| "grad_norm": 12.5625, |
| "learning_rate": 1.800740740740741e-05, |
| "loss": 15.6226, |
| "mean_token_accuracy": 0.7802051860839129, |
| "num_tokens": 18869344.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.9295620009303093, |
| "epoch": 0.928, |
| "grad_norm": 12.75, |
| "learning_rate": 1.7933333333333333e-05, |
| "loss": 15.5339, |
| "mean_token_accuracy": 0.7815026968717576, |
| "num_tokens": 19199834.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.9112746389582753, |
| "epoch": 0.944, |
| "grad_norm": 12.5, |
| "learning_rate": 1.785925925925926e-05, |
| "loss": 15.2153, |
| "mean_token_accuracy": 0.7855059750378132, |
| "num_tokens": 19525385.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.9181267462670804, |
| "epoch": 0.96, |
| "grad_norm": 12.75, |
| "learning_rate": 1.7785185185185186e-05, |
| "loss": 15.3751, |
| "mean_token_accuracy": 0.7829241104424, |
| "num_tokens": 19856973.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 2.594178828239441, |
| "eval_biology_loss": 2.9207751750946045, |
| "eval_biology_mean_token_accuracy": 0.5271391036510468, |
| "eval_biology_num_tokens": 19856973.0, |
| "eval_biology_runtime": 57.0207, |
| "eval_biology_samples_per_second": 8.769, |
| "eval_biology_steps_per_second": 2.192, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_chemistry_entropy": 1.270768217086792, |
| "eval_chemistry_loss": 1.3242253065109253, |
| "eval_chemistry_mean_token_accuracy": 0.7280859136581421, |
| "eval_chemistry_num_tokens": 19856973.0, |
| "eval_chemistry_runtime": 72.006, |
| "eval_chemistry_samples_per_second": 6.944, |
| "eval_chemistry_steps_per_second": 1.736, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_math_entropy": 0.9325860266685486, |
| "eval_math_loss": 1.1655522584915161, |
| "eval_math_mean_token_accuracy": 0.7533689908981324, |
| "eval_math_num_tokens": 19856973.0, |
| "eval_math_runtime": 74.0585, |
| "eval_math_samples_per_second": 6.751, |
| "eval_math_steps_per_second": 1.688, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_physics_entropy": 0.9269161529541016, |
| "eval_physics_loss": 0.9667912125587463, |
| "eval_physics_mean_token_accuracy": 0.7814396271705627, |
| "eval_physics_num_tokens": 19856973.0, |
| "eval_physics_runtime": 84.5863, |
| "eval_physics_samples_per_second": 5.911, |
| "eval_physics_steps_per_second": 1.478, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.9162795972079039, |
| "epoch": 0.976, |
| "grad_norm": 12.125, |
| "learning_rate": 1.7711111111111113e-05, |
| "loss": 15.2813, |
| "mean_token_accuracy": 0.7860829800367355, |
| "num_tokens": 20190600.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.9235894799232482, |
| "epoch": 0.992, |
| "grad_norm": 11.5, |
| "learning_rate": 1.763703703703704e-05, |
| "loss": 15.4241, |
| "mean_token_accuracy": 0.784028148651123, |
| "num_tokens": 20518014.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.9041305113583803, |
| "epoch": 1.008, |
| "grad_norm": 11.0625, |
| "learning_rate": 1.7562962962962962e-05, |
| "loss": 14.9099, |
| "mean_token_accuracy": 0.7884359251707792, |
| "num_tokens": 20859215.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.9018714495003224, |
| "epoch": 1.024, |
| "grad_norm": 11.1875, |
| "learning_rate": 1.7488888888888892e-05, |
| "loss": 15.1658, |
| "mean_token_accuracy": 0.7857498530298471, |
| "num_tokens": 21186658.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.916073745302856, |
| "epoch": 1.04, |
| "grad_norm": 10.8125, |
| "learning_rate": 1.7414814814814815e-05, |
| "loss": 15.1367, |
| "mean_token_accuracy": 0.7867844242602586, |
| "num_tokens": 21517237.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.8802109774202108, |
| "epoch": 1.056, |
| "grad_norm": 11.0, |
| "learning_rate": 1.7340740740740742e-05, |
| "loss": 14.7961, |
| "mean_token_accuracy": 0.7896511305123568, |
| "num_tokens": 21852528.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.9102928217500448, |
| "epoch": 1.072, |
| "grad_norm": 10.0, |
| "learning_rate": 1.726666666666667e-05, |
| "loss": 15.1014, |
| "mean_token_accuracy": 0.7862772591412067, |
| "num_tokens": 22192380.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.9107371259480714, |
| "epoch": 1.088, |
| "grad_norm": 13.4375, |
| "learning_rate": 1.7192592592592595e-05, |
| "loss": 15.2493, |
| "mean_token_accuracy": 0.7848471954464913, |
| "num_tokens": 22517971.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.8938208010047675, |
| "epoch": 1.104, |
| "grad_norm": 10.6875, |
| "learning_rate": 1.711851851851852e-05, |
| "loss": 14.9017, |
| "mean_token_accuracy": 0.7887144055217504, |
| "num_tokens": 22848694.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.9129897281527519, |
| "epoch": 1.12, |
| "grad_norm": 11.4375, |
| "learning_rate": 1.7044444444444445e-05, |
| "loss": 15.2722, |
| "mean_token_accuracy": 0.7840389590710402, |
| "num_tokens": 23183889.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_biology_entropy": 2.5242818908691405, |
| "eval_biology_loss": 2.9092111587524414, |
| "eval_biology_mean_token_accuracy": 0.5278490686416626, |
| "eval_biology_num_tokens": 23183889.0, |
| "eval_biology_runtime": 57.2117, |
| "eval_biology_samples_per_second": 8.739, |
| "eval_biology_steps_per_second": 2.185, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_chemistry_entropy": 1.2367294158935547, |
| "eval_chemistry_loss": 1.3155030012130737, |
| "eval_chemistry_mean_token_accuracy": 0.7286214451789856, |
| "eval_chemistry_num_tokens": 23183889.0, |
| "eval_chemistry_runtime": 71.8224, |
| "eval_chemistry_samples_per_second": 6.962, |
| "eval_chemistry_steps_per_second": 1.74, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_math_entropy": 0.9127886519432068, |
| "eval_math_loss": 1.1603002548217773, |
| "eval_math_mean_token_accuracy": 0.754603410243988, |
| "eval_math_num_tokens": 23183889.0, |
| "eval_math_runtime": 74.5403, |
| "eval_math_samples_per_second": 6.708, |
| "eval_math_steps_per_second": 1.677, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_physics_entropy": 0.9048804240226745, |
| "eval_physics_loss": 0.9561019539833069, |
| "eval_physics_mean_token_accuracy": 0.7832872924804688, |
| "eval_physics_num_tokens": 23183889.0, |
| "eval_physics_runtime": 84.6107, |
| "eval_physics_samples_per_second": 5.909, |
| "eval_physics_steps_per_second": 1.477, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.8741975981742144, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 10.1875, |
| "learning_rate": 1.697037037037037e-05, |
| "loss": 14.564, |
| "mean_token_accuracy": 0.7930445209145546, |
| "num_tokens": 23523903.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.8917050797492265, |
| "epoch": 1.152, |
| "grad_norm": 12.75, |
| "learning_rate": 1.6896296296296298e-05, |
| "loss": 14.9557, |
| "mean_token_accuracy": 0.7896463591605425, |
| "num_tokens": 23848848.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.9238279532641173, |
| "epoch": 1.168, |
| "grad_norm": 11.6875, |
| "learning_rate": 1.6822222222222224e-05, |
| "loss": 15.2756, |
| "mean_token_accuracy": 0.7821665916591882, |
| "num_tokens": 24174589.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.9052068604156375, |
| "epoch": 1.184, |
| "grad_norm": 13.5625, |
| "learning_rate": 1.6748148148148147e-05, |
| "loss": 15.0924, |
| "mean_token_accuracy": 0.7870742384344339, |
| "num_tokens": 24495981.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.877324876561761, |
| "epoch": 1.2, |
| "grad_norm": 11.1875, |
| "learning_rate": 1.6674074074074077e-05, |
| "loss": 14.7224, |
| "mean_token_accuracy": 0.7919886518269778, |
| "num_tokens": 24826245.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.9162866469472647, |
| "epoch": 1.216, |
| "grad_norm": 14.1875, |
| "learning_rate": 1.66e-05, |
| "loss": 15.1252, |
| "mean_token_accuracy": 0.7850615303963423, |
| "num_tokens": 25147775.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.8838450819253921, |
| "epoch": 1.232, |
| "grad_norm": 12.1875, |
| "learning_rate": 1.6525925925925927e-05, |
| "loss": 14.8829, |
| "mean_token_accuracy": 0.7893829133361578, |
| "num_tokens": 25480033.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.8952008984982968, |
| "epoch": 1.248, |
| "grad_norm": 12.875, |
| "learning_rate": 1.6451851851851853e-05, |
| "loss": 14.9213, |
| "mean_token_accuracy": 0.7889137178659439, |
| "num_tokens": 25811978.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.8893922835588455, |
| "epoch": 1.264, |
| "grad_norm": 12.6875, |
| "learning_rate": 1.637777777777778e-05, |
| "loss": 14.7989, |
| "mean_token_accuracy": 0.7897166382521391, |
| "num_tokens": 26146299.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.9172267891466618, |
| "epoch": 1.28, |
| "grad_norm": 11.75, |
| "learning_rate": 1.6303703703703706e-05, |
| "loss": 15.3289, |
| "mean_token_accuracy": 0.782142074778676, |
| "num_tokens": 26478207.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_biology_entropy": 2.5403532466888428, |
| "eval_biology_loss": 2.9071850776672363, |
| "eval_biology_mean_token_accuracy": 0.529075275182724, |
| "eval_biology_num_tokens": 26478207.0, |
| "eval_biology_runtime": 56.934, |
| "eval_biology_samples_per_second": 8.782, |
| "eval_biology_steps_per_second": 2.196, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_chemistry_entropy": 1.234187596321106, |
| "eval_chemistry_loss": 1.306308388710022, |
| "eval_chemistry_mean_token_accuracy": 0.7299280157089233, |
| "eval_chemistry_num_tokens": 26478207.0, |
| "eval_chemistry_runtime": 71.938, |
| "eval_chemistry_samples_per_second": 6.95, |
| "eval_chemistry_steps_per_second": 1.738, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_math_entropy": 0.9124487805366516, |
| "eval_math_loss": 1.1551406383514404, |
| "eval_math_mean_token_accuracy": 0.7556050124168396, |
| "eval_math_num_tokens": 26478207.0, |
| "eval_math_runtime": 75.1703, |
| "eval_math_samples_per_second": 6.652, |
| "eval_math_steps_per_second": 1.663, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_physics_entropy": 0.8998128218650818, |
| "eval_physics_loss": 0.9489150047302246, |
| "eval_physics_mean_token_accuracy": 0.7840672006607056, |
| "eval_physics_num_tokens": 26478207.0, |
| "eval_physics_runtime": 85.0147, |
| "eval_physics_samples_per_second": 5.881, |
| "eval_physics_steps_per_second": 1.47, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.8760543465614319, |
| "epoch": 1.296, |
| "grad_norm": 12.8125, |
| "learning_rate": 1.622962962962963e-05, |
| "loss": 14.6611, |
| "mean_token_accuracy": 0.7923007346689701, |
| "num_tokens": 26806477.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.917656009271741, |
| "epoch": 1.312, |
| "grad_norm": 13.1875, |
| "learning_rate": 1.6155555555555556e-05, |
| "loss": 15.2379, |
| "mean_token_accuracy": 0.7841715186834335, |
| "num_tokens": 27137528.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.8967037841677665, |
| "epoch": 1.328, |
| "grad_norm": 13.25, |
| "learning_rate": 1.6081481481481482e-05, |
| "loss": 14.9513, |
| "mean_token_accuracy": 0.7878007985651493, |
| "num_tokens": 27465859.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.901091538183391, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 13.375, |
| "learning_rate": 1.600740740740741e-05, |
| "loss": 15.1044, |
| "mean_token_accuracy": 0.7861052300781012, |
| "num_tokens": 27795568.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.9121165912598371, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 11.625, |
| "learning_rate": 1.5933333333333336e-05, |
| "loss": 15.0779, |
| "mean_token_accuracy": 0.7859285034239292, |
| "num_tokens": 28129988.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.8819791020825505, |
| "epoch": 1.376, |
| "grad_norm": 13.875, |
| "learning_rate": 1.5859259259259262e-05, |
| "loss": 14.7997, |
| "mean_token_accuracy": 0.7886631991714239, |
| "num_tokens": 28465903.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.8854727063328027, |
| "epoch": 1.392, |
| "grad_norm": 13.25, |
| "learning_rate": 1.5785185185185185e-05, |
| "loss": 14.61, |
| "mean_token_accuracy": 0.7927087835967541, |
| "num_tokens": 28801554.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.8729526123031974, |
| "epoch": 1.408, |
| "grad_norm": 12.4375, |
| "learning_rate": 1.571111111111111e-05, |
| "loss": 14.7562, |
| "mean_token_accuracy": 0.7899845536798239, |
| "num_tokens": 29125114.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.913954821228981, |
| "epoch": 1.424, |
| "grad_norm": 10.9375, |
| "learning_rate": 1.5637037037037038e-05, |
| "loss": 15.2188, |
| "mean_token_accuracy": 0.783445942774415, |
| "num_tokens": 29454334.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.8652864851057529, |
| "epoch": 1.44, |
| "grad_norm": 12.1875, |
| "learning_rate": 1.5562962962962965e-05, |
| "loss": 14.3528, |
| "mean_token_accuracy": 0.794481047987938, |
| "num_tokens": 29789253.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_biology_entropy": 2.4760996789932253, |
| "eval_biology_loss": 2.897955894470215, |
| "eval_biology_mean_token_accuracy": 0.5296812095642089, |
| "eval_biology_num_tokens": 29789253.0, |
| "eval_biology_runtime": 57.0842, |
| "eval_biology_samples_per_second": 8.759, |
| "eval_biology_steps_per_second": 2.19, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_chemistry_entropy": 1.2120891718864442, |
| "eval_chemistry_loss": 1.2991372346878052, |
| "eval_chemistry_mean_token_accuracy": 0.7307901701927185, |
| "eval_chemistry_num_tokens": 29789253.0, |
| "eval_chemistry_runtime": 71.5779, |
| "eval_chemistry_samples_per_second": 6.985, |
| "eval_chemistry_steps_per_second": 1.746, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_math_entropy": 0.9021294908523559, |
| "eval_math_loss": 1.1532059907913208, |
| "eval_math_mean_token_accuracy": 0.755294855594635, |
| "eval_math_num_tokens": 29789253.0, |
| "eval_math_runtime": 74.8641, |
| "eval_math_samples_per_second": 6.679, |
| "eval_math_steps_per_second": 1.67, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_physics_entropy": 0.8853664617538453, |
| "eval_physics_loss": 0.9413084983825684, |
| "eval_physics_mean_token_accuracy": 0.7854379539489746, |
| "eval_physics_num_tokens": 29789253.0, |
| "eval_physics_runtime": 84.3345, |
| "eval_physics_samples_per_second": 5.929, |
| "eval_physics_steps_per_second": 1.482, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.8592871313914656, |
| "epoch": 1.456, |
| "grad_norm": 10.5, |
| "learning_rate": 1.548888888888889e-05, |
| "loss": 14.4092, |
| "mean_token_accuracy": 0.7947866775095462, |
| "num_tokens": 30128576.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.9165356293320656, |
| "epoch": 1.472, |
| "grad_norm": 12.8125, |
| "learning_rate": 1.5414814814814814e-05, |
| "loss": 15.368, |
| "mean_token_accuracy": 0.7829437594860792, |
| "num_tokens": 30452719.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.9012311644852161, |
| "epoch": 1.488, |
| "grad_norm": 10.9375, |
| "learning_rate": 1.5340740740740744e-05, |
| "loss": 14.9137, |
| "mean_token_accuracy": 0.7879349775612354, |
| "num_tokens": 30782343.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.8793585889041424, |
| "epoch": 1.504, |
| "grad_norm": 11.75, |
| "learning_rate": 1.5266666666666667e-05, |
| "loss": 14.6612, |
| "mean_token_accuracy": 0.7913552910089493, |
| "num_tokens": 31113238.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.896404268220067, |
| "epoch": 1.52, |
| "grad_norm": 12.25, |
| "learning_rate": 1.5192592592592594e-05, |
| "loss": 15.0177, |
| "mean_token_accuracy": 0.7869206938892603, |
| "num_tokens": 31444589.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.9115363927558064, |
| "epoch": 1.536, |
| "grad_norm": 13.625, |
| "learning_rate": 1.5118518518518519e-05, |
| "loss": 15.1508, |
| "mean_token_accuracy": 0.7850655887275935, |
| "num_tokens": 31774137.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.8715274870395661, |
| "epoch": 1.552, |
| "grad_norm": 12.5, |
| "learning_rate": 1.5044444444444445e-05, |
| "loss": 14.4724, |
| "mean_token_accuracy": 0.7919801536947488, |
| "num_tokens": 32101776.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.8716113748028874, |
| "epoch": 1.568, |
| "grad_norm": 11.9375, |
| "learning_rate": 1.497037037037037e-05, |
| "loss": 14.5441, |
| "mean_token_accuracy": 0.7928904049098492, |
| "num_tokens": 32428021.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.9078653456643224, |
| "epoch": 1.584, |
| "grad_norm": 12.375, |
| "learning_rate": 1.4896296296296298e-05, |
| "loss": 15.1365, |
| "mean_token_accuracy": 0.7834811493754387, |
| "num_tokens": 32752735.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.8948801588267088, |
| "epoch": 1.6, |
| "grad_norm": 11.4375, |
| "learning_rate": 1.4822222222222225e-05, |
| "loss": 14.975, |
| "mean_token_accuracy": 0.7872815534472466, |
| "num_tokens": 33075822.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_biology_entropy": 2.508749726295471, |
| "eval_biology_loss": 2.8778915405273438, |
| "eval_biology_mean_token_accuracy": 0.5310802006721497, |
| "eval_biology_num_tokens": 33075822.0, |
| "eval_biology_runtime": 57.246, |
| "eval_biology_samples_per_second": 8.734, |
| "eval_biology_steps_per_second": 2.184, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_chemistry_entropy": 1.2282180700302123, |
| "eval_chemistry_loss": 1.2942099571228027, |
| "eval_chemistry_mean_token_accuracy": 0.7318216099739074, |
| "eval_chemistry_num_tokens": 33075822.0, |
| "eval_chemistry_runtime": 71.0784, |
| "eval_chemistry_samples_per_second": 7.034, |
| "eval_chemistry_steps_per_second": 1.759, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_math_entropy": 0.9116221733093262, |
| "eval_math_loss": 1.1501303911209106, |
| "eval_math_mean_token_accuracy": 0.7557219748497009, |
| "eval_math_num_tokens": 33075822.0, |
| "eval_math_runtime": 74.8337, |
| "eval_math_samples_per_second": 6.681, |
| "eval_math_steps_per_second": 1.67, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_physics_entropy": 0.8932713055610657, |
| "eval_physics_loss": 0.9365818500518799, |
| "eval_physics_mean_token_accuracy": 0.785537106513977, |
| "eval_physics_num_tokens": 33075822.0, |
| "eval_physics_runtime": 84.6443, |
| "eval_physics_samples_per_second": 5.907, |
| "eval_physics_steps_per_second": 1.477, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.8604803189635277, |
| "epoch": 1.616, |
| "grad_norm": 12.5, |
| "learning_rate": 1.474814814814815e-05, |
| "loss": 14.3639, |
| "mean_token_accuracy": 0.7951964545994997, |
| "num_tokens": 33411850.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.8778028018772602, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 12.5625, |
| "learning_rate": 1.4674074074074076e-05, |
| "loss": 14.6452, |
| "mean_token_accuracy": 0.7921328764408827, |
| "num_tokens": 33739456.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.853967827372253, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 12.375, |
| "learning_rate": 1.46e-05, |
| "loss": 14.2504, |
| "mean_token_accuracy": 0.7954076964408159, |
| "num_tokens": 34072022.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.8785163436084986, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 13.5625, |
| "learning_rate": 1.4525925925925927e-05, |
| "loss": 14.7032, |
| "mean_token_accuracy": 0.7915361914783716, |
| "num_tokens": 34398959.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.8694784231483936, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 11.1875, |
| "learning_rate": 1.4451851851851852e-05, |
| "loss": 14.3839, |
| "mean_token_accuracy": 0.7940326742827892, |
| "num_tokens": 34732284.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.8717694684863091, |
| "epoch": 1.696, |
| "grad_norm": 13.1875, |
| "learning_rate": 1.4377777777777779e-05, |
| "loss": 14.6494, |
| "mean_token_accuracy": 0.7915462471544743, |
| "num_tokens": 35053712.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.8986424550414085, |
| "epoch": 1.712, |
| "grad_norm": 14.4375, |
| "learning_rate": 1.4303703703703703e-05, |
| "loss": 14.8976, |
| "mean_token_accuracy": 0.7876383919268847, |
| "num_tokens": 35393088.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.8717451239004731, |
| "epoch": 1.728, |
| "grad_norm": 12.6875, |
| "learning_rate": 1.4229629629629632e-05, |
| "loss": 14.5646, |
| "mean_token_accuracy": 0.7918220609426498, |
| "num_tokens": 35729007.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.8697011994197965, |
| "epoch": 1.744, |
| "grad_norm": 13.5, |
| "learning_rate": 1.4155555555555556e-05, |
| "loss": 14.5182, |
| "mean_token_accuracy": 0.7922107793390751, |
| "num_tokens": 36072489.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.8740223359316588, |
| "epoch": 1.76, |
| "grad_norm": 12.0, |
| "learning_rate": 1.4081481481481483e-05, |
| "loss": 14.4753, |
| "mean_token_accuracy": 0.7926081418991089, |
| "num_tokens": 36398285.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_biology_entropy": 2.487686212539673, |
| "eval_biology_loss": 2.874035120010376, |
| "eval_biology_mean_token_accuracy": 0.5334258432388306, |
| "eval_biology_num_tokens": 36398285.0, |
| "eval_biology_runtime": 57.5039, |
| "eval_biology_samples_per_second": 8.695, |
| "eval_biology_steps_per_second": 2.174, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_chemistry_entropy": 1.2144387559890748, |
| "eval_chemistry_loss": 1.29018235206604, |
| "eval_chemistry_mean_token_accuracy": 0.7329304647445679, |
| "eval_chemistry_num_tokens": 36398285.0, |
| "eval_chemistry_runtime": 71.5726, |
| "eval_chemistry_samples_per_second": 6.986, |
| "eval_chemistry_steps_per_second": 1.746, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_math_entropy": 0.9051191382408142, |
| "eval_math_loss": 1.1472207307815552, |
| "eval_math_mean_token_accuracy": 0.7566594734191895, |
| "eval_math_num_tokens": 36398285.0, |
| "eval_math_runtime": 74.0714, |
| "eval_math_samples_per_second": 6.75, |
| "eval_math_steps_per_second": 1.688, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_physics_entropy": 0.8845951933860778, |
| "eval_physics_loss": 0.9326092600822449, |
| "eval_physics_mean_token_accuracy": 0.7863762426376343, |
| "eval_physics_num_tokens": 36398285.0, |
| "eval_physics_runtime": 84.7278, |
| "eval_physics_samples_per_second": 5.901, |
| "eval_physics_steps_per_second": 1.475, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.9020034002140165, |
| "epoch": 1.776, |
| "grad_norm": 12.625, |
| "learning_rate": 1.400740740740741e-05, |
| "loss": 15.0724, |
| "mean_token_accuracy": 0.7856271650642157, |
| "num_tokens": 36730479.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.862918908149004, |
| "epoch": 1.792, |
| "grad_norm": 14.6875, |
| "learning_rate": 1.3933333333333334e-05, |
| "loss": 14.5005, |
| "mean_token_accuracy": 0.7926642511039972, |
| "num_tokens": 37057425.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.8617109794169664, |
| "epoch": 1.808, |
| "grad_norm": 11.8125, |
| "learning_rate": 1.385925925925926e-05, |
| "loss": 14.2625, |
| "mean_token_accuracy": 0.7952517606317997, |
| "num_tokens": 37395449.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.8684184370562434, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 12.5, |
| "learning_rate": 1.3785185185185186e-05, |
| "loss": 14.5241, |
| "mean_token_accuracy": 0.7931536667048931, |
| "num_tokens": 37729535.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.8654272655025125, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 12.4375, |
| "learning_rate": 1.3711111111111112e-05, |
| "loss": 14.4054, |
| "mean_token_accuracy": 0.7938401244580746, |
| "num_tokens": 38057839.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.86546247061342, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 13.4375, |
| "learning_rate": 1.3637037037037037e-05, |
| "loss": 14.4546, |
| "mean_token_accuracy": 0.7931617993861437, |
| "num_tokens": 38398033.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.8690170094370842, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 11.8125, |
| "learning_rate": 1.3562962962962965e-05, |
| "loss": 14.4475, |
| "mean_token_accuracy": 0.7915640283375979, |
| "num_tokens": 38729203.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.8866201037541032, |
| "epoch": 1.888, |
| "grad_norm": 13.0625, |
| "learning_rate": 1.3488888888888888e-05, |
| "loss": 14.8192, |
| "mean_token_accuracy": 0.7894200544804335, |
| "num_tokens": 39058791.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.8984544143080712, |
| "epoch": 1.904, |
| "grad_norm": 12.25, |
| "learning_rate": 1.3414814814814817e-05, |
| "loss": 15.0873, |
| "mean_token_accuracy": 0.7861377280205488, |
| "num_tokens": 39386170.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.9040035644546152, |
| "epoch": 1.92, |
| "grad_norm": 12.0625, |
| "learning_rate": 1.3340740740740741e-05, |
| "loss": 14.9941, |
| "mean_token_accuracy": 0.7868796121329069, |
| "num_tokens": 39710381.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_biology_entropy": 2.485040246009827, |
| "eval_biology_loss": 2.8711678981781006, |
| "eval_biology_mean_token_accuracy": 0.5342430763244629, |
| "eval_biology_num_tokens": 39710381.0, |
| "eval_biology_runtime": 57.5639, |
| "eval_biology_samples_per_second": 8.686, |
| "eval_biology_steps_per_second": 2.171, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_chemistry_entropy": 1.208739689350128, |
| "eval_chemistry_loss": 1.285979151725769, |
| "eval_chemistry_mean_token_accuracy": 0.7334310331344605, |
| "eval_chemistry_num_tokens": 39710381.0, |
| "eval_chemistry_runtime": 71.5821, |
| "eval_chemistry_samples_per_second": 6.985, |
| "eval_chemistry_steps_per_second": 1.746, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_math_entropy": 0.9018509998321533, |
| "eval_math_loss": 1.1460354328155518, |
| "eval_math_mean_token_accuracy": 0.7567954559326172, |
| "eval_math_num_tokens": 39710381.0, |
| "eval_math_runtime": 75.1131, |
| "eval_math_samples_per_second": 6.657, |
| "eval_math_steps_per_second": 1.664, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_physics_entropy": 0.8798905625343323, |
| "eval_physics_loss": 0.929322361946106, |
| "eval_physics_mean_token_accuracy": 0.787430817604065, |
| "eval_physics_num_tokens": 39710381.0, |
| "eval_physics_runtime": 84.784, |
| "eval_physics_samples_per_second": 5.897, |
| "eval_physics_steps_per_second": 1.474, |
| "step": 1200 |
| }, |
| { |
| "entropy": 0.8708227036520839, |
| "epoch": 1.936, |
| "grad_norm": 13.8125, |
| "learning_rate": 1.3266666666666668e-05, |
| "loss": 14.626, |
| "mean_token_accuracy": 0.7907645083963871, |
| "num_tokens": 40043354.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 0.8374214060604572, |
| "epoch": 1.952, |
| "grad_norm": 11.5, |
| "learning_rate": 1.3192592592592594e-05, |
| "loss": 13.9343, |
| "mean_token_accuracy": 0.7994464814662934, |
| "num_tokens": 40375790.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 0.8776708476245403, |
| "epoch": 1.968, |
| "grad_norm": 12.25, |
| "learning_rate": 1.311851851851852e-05, |
| "loss": 14.5653, |
| "mean_token_accuracy": 0.7916903175413609, |
| "num_tokens": 40705451.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 0.8854383405297994, |
| "epoch": 1.984, |
| "grad_norm": 13.0, |
| "learning_rate": 1.3044444444444446e-05, |
| "loss": 14.7486, |
| "mean_token_accuracy": 0.7898764718323946, |
| "num_tokens": 41040313.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 0.879773647710681, |
| "epoch": 2.0, |
| "grad_norm": 12.6875, |
| "learning_rate": 1.297037037037037e-05, |
| "loss": 14.6916, |
| "mean_token_accuracy": 0.7911134500056505, |
| "num_tokens": 41373608.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.8601289037615061, |
| "epoch": 2.016, |
| "grad_norm": 10.5625, |
| "learning_rate": 1.2896296296296299e-05, |
| "loss": 14.1908, |
| "mean_token_accuracy": 0.7955393616110087, |
| "num_tokens": 41708226.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.8542119370773434, |
| "epoch": 2.032, |
| "grad_norm": 11.4375, |
| "learning_rate": 1.2822222222222222e-05, |
| "loss": 14.2867, |
| "mean_token_accuracy": 0.7959718242287636, |
| "num_tokens": 42045433.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.8921715309843421, |
| "epoch": 2.048, |
| "grad_norm": 13.1875, |
| "learning_rate": 1.274814814814815e-05, |
| "loss": 14.8772, |
| "mean_token_accuracy": 0.7874497707933188, |
| "num_tokens": 42372694.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.8619997894391418, |
| "epoch": 2.064, |
| "grad_norm": 12.375, |
| "learning_rate": 1.2674074074074075e-05, |
| "loss": 14.2556, |
| "mean_token_accuracy": 0.7947564154863358, |
| "num_tokens": 42710240.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.8751549379900098, |
| "epoch": 2.08, |
| "grad_norm": 11.5625, |
| "learning_rate": 1.2600000000000001e-05, |
| "loss": 14.6573, |
| "mean_token_accuracy": 0.7916811484843492, |
| "num_tokens": 43037844.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_biology_entropy": 2.5093424930572508, |
| "eval_biology_loss": 2.864481210708618, |
| "eval_biology_mean_token_accuracy": 0.533148678779602, |
| "eval_biology_num_tokens": 43037844.0, |
| "eval_biology_runtime": 57.534, |
| "eval_biology_samples_per_second": 8.691, |
| "eval_biology_steps_per_second": 2.173, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_chemistry_entropy": 1.2106250019073486, |
| "eval_chemistry_loss": 1.2846547365188599, |
| "eval_chemistry_mean_token_accuracy": 0.7332292509078979, |
| "eval_chemistry_num_tokens": 43037844.0, |
| "eval_chemistry_runtime": 71.3949, |
| "eval_chemistry_samples_per_second": 7.003, |
| "eval_chemistry_steps_per_second": 1.751, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_math_entropy": 0.9025118923187256, |
| "eval_math_loss": 1.1455634832382202, |
| "eval_math_mean_token_accuracy": 0.7569516344070435, |
| "eval_math_num_tokens": 43037844.0, |
| "eval_math_runtime": 75.2696, |
| "eval_math_samples_per_second": 6.643, |
| "eval_math_steps_per_second": 1.661, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_physics_entropy": 0.8776944227218628, |
| "eval_physics_loss": 0.9265705943107605, |
| "eval_physics_mean_token_accuracy": 0.7875892186164856, |
| "eval_physics_num_tokens": 43037844.0, |
| "eval_physics_runtime": 85.0265, |
| "eval_physics_samples_per_second": 5.881, |
| "eval_physics_steps_per_second": 1.47, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.8755308095365762, |
| "epoch": 2.096, |
| "grad_norm": 12.9375, |
| "learning_rate": 1.2525925925925928e-05, |
| "loss": 14.5668, |
| "mean_token_accuracy": 0.7918181721121073, |
| "num_tokens": 43368675.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.8589686010032892, |
| "epoch": 2.112, |
| "grad_norm": 12.25, |
| "learning_rate": 1.2451851851851853e-05, |
| "loss": 14.3302, |
| "mean_token_accuracy": 0.7948925741016865, |
| "num_tokens": 43699544.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.8743958847597242, |
| "epoch": 2.128, |
| "grad_norm": 12.3125, |
| "learning_rate": 1.237777777777778e-05, |
| "loss": 14.5493, |
| "mean_token_accuracy": 0.7907001547515392, |
| "num_tokens": 44012632.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.8594208642840385, |
| "epoch": 2.144, |
| "grad_norm": 11.0625, |
| "learning_rate": 1.2303703703703704e-05, |
| "loss": 14.3324, |
| "mean_token_accuracy": 0.7929412983357906, |
| "num_tokens": 44349732.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.8937922870740295, |
| "epoch": 2.16, |
| "grad_norm": 12.625, |
| "learning_rate": 1.222962962962963e-05, |
| "loss": 14.8834, |
| "mean_token_accuracy": 0.7873219456523657, |
| "num_tokens": 44680410.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.8523136384785175, |
| "epoch": 2.176, |
| "grad_norm": 11.25, |
| "learning_rate": 1.2155555555555555e-05, |
| "loss": 14.2544, |
| "mean_token_accuracy": 0.7959212839603425, |
| "num_tokens": 45017922.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.8626351818442345, |
| "epoch": 2.192, |
| "grad_norm": 12.1875, |
| "learning_rate": 1.2081481481481484e-05, |
| "loss": 14.276, |
| "mean_token_accuracy": 0.7951519623398781, |
| "num_tokens": 45348179.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.8525467690080404, |
| "epoch": 2.208, |
| "grad_norm": 12.8125, |
| "learning_rate": 1.2007407407407408e-05, |
| "loss": 14.2013, |
| "mean_token_accuracy": 0.7966051142662763, |
| "num_tokens": 45678397.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.8527483340352774, |
| "epoch": 2.224, |
| "grad_norm": 11.5625, |
| "learning_rate": 1.1933333333333335e-05, |
| "loss": 14.2627, |
| "mean_token_accuracy": 0.7949573740363121, |
| "num_tokens": 46010052.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.855360858142376, |
| "epoch": 2.24, |
| "grad_norm": 11.4375, |
| "learning_rate": 1.185925925925926e-05, |
| "loss": 14.2457, |
| "mean_token_accuracy": 0.7951466862112284, |
| "num_tokens": 46347116.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_biology_entropy": 2.5089185886383056, |
| "eval_biology_loss": 2.863374948501587, |
| "eval_biology_mean_token_accuracy": 0.5349591159820557, |
| "eval_biology_num_tokens": 46347116.0, |
| "eval_biology_runtime": 57.3344, |
| "eval_biology_samples_per_second": 8.721, |
| "eval_biology_steps_per_second": 2.18, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_chemistry_entropy": 1.2185512976646424, |
| "eval_chemistry_loss": 1.2818259000778198, |
| "eval_chemistry_mean_token_accuracy": 0.7342271280288696, |
| "eval_chemistry_num_tokens": 46347116.0, |
| "eval_chemistry_runtime": 71.4639, |
| "eval_chemistry_samples_per_second": 6.997, |
| "eval_chemistry_steps_per_second": 1.749, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_math_entropy": 0.9067861828804016, |
| "eval_math_loss": 1.1431993246078491, |
| "eval_math_mean_token_accuracy": 0.7572210087776184, |
| "eval_math_num_tokens": 46347116.0, |
| "eval_math_runtime": 75.3577, |
| "eval_math_samples_per_second": 6.635, |
| "eval_math_steps_per_second": 1.659, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_physics_entropy": 0.8840774774551392, |
| "eval_physics_loss": 0.9245942234992981, |
| "eval_physics_mean_token_accuracy": 0.7878187990188599, |
| "eval_physics_num_tokens": 46347116.0, |
| "eval_physics_runtime": 84.9011, |
| "eval_physics_samples_per_second": 5.889, |
| "eval_physics_steps_per_second": 1.472, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.9004593985155225, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 13.4375, |
| "learning_rate": 1.1785185185185186e-05, |
| "loss": 14.9385, |
| "mean_token_accuracy": 0.7881197012960911, |
| "num_tokens": 46673074.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.8454325065016747, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 12.125, |
| "learning_rate": 1.1711111111111113e-05, |
| "loss": 14.0933, |
| "mean_token_accuracy": 0.7979227758944034, |
| "num_tokens": 47009386.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.8827449329197407, |
| "epoch": 2.288, |
| "grad_norm": 12.25, |
| "learning_rate": 1.1637037037037037e-05, |
| "loss": 14.712, |
| "mean_token_accuracy": 0.7902342595160008, |
| "num_tokens": 47344231.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.8806605780497193, |
| "epoch": 2.304, |
| "grad_norm": 13.4375, |
| "learning_rate": 1.1562962962962964e-05, |
| "loss": 14.5674, |
| "mean_token_accuracy": 0.7924861270934344, |
| "num_tokens": 47658392.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.859540069103241, |
| "epoch": 2.32, |
| "grad_norm": 12.1875, |
| "learning_rate": 1.1488888888888889e-05, |
| "loss": 14.3445, |
| "mean_token_accuracy": 0.7936220798641443, |
| "num_tokens": 47995015.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.8417845372110605, |
| "epoch": 2.336, |
| "grad_norm": 11.8125, |
| "learning_rate": 1.1414814814814817e-05, |
| "loss": 14.0115, |
| "mean_token_accuracy": 0.7978896964341402, |
| "num_tokens": 48336708.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.8409199349582195, |
| "epoch": 2.352, |
| "grad_norm": 11.5, |
| "learning_rate": 1.1340740740740742e-05, |
| "loss": 13.9932, |
| "mean_token_accuracy": 0.7971417736262083, |
| "num_tokens": 48665311.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.8575670935213566, |
| "epoch": 2.368, |
| "grad_norm": 11.75, |
| "learning_rate": 1.1266666666666668e-05, |
| "loss": 14.3902, |
| "mean_token_accuracy": 0.7932044431567192, |
| "num_tokens": 49005334.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.8846392493695021, |
| "epoch": 2.384, |
| "grad_norm": 13.5, |
| "learning_rate": 1.1192592592592593e-05, |
| "loss": 14.5912, |
| "mean_token_accuracy": 0.7901103302836419, |
| "num_tokens": 49331936.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.8669425016269088, |
| "epoch": 2.4, |
| "grad_norm": 12.4375, |
| "learning_rate": 1.111851851851852e-05, |
| "loss": 14.4765, |
| "mean_token_accuracy": 0.7921019058674574, |
| "num_tokens": 49671048.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_biology_entropy": 2.456291440010071, |
| "eval_biology_loss": 2.8605849742889404, |
| "eval_biology_mean_token_accuracy": 0.5338850898742675, |
| "eval_biology_num_tokens": 49671048.0, |
| "eval_biology_runtime": 57.1011, |
| "eval_biology_samples_per_second": 8.756, |
| "eval_biology_steps_per_second": 2.189, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_chemistry_entropy": 1.1939663624763488, |
| "eval_chemistry_loss": 1.280471682548523, |
| "eval_chemistry_mean_token_accuracy": 0.7337013077735901, |
| "eval_chemistry_num_tokens": 49671048.0, |
| "eval_chemistry_runtime": 71.0078, |
| "eval_chemistry_samples_per_second": 7.041, |
| "eval_chemistry_steps_per_second": 1.76, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_math_entropy": 0.8943547616004944, |
| "eval_math_loss": 1.1426331996917725, |
| "eval_math_mean_token_accuracy": 0.7575869603157044, |
| "eval_math_num_tokens": 49671048.0, |
| "eval_math_runtime": 74.2145, |
| "eval_math_samples_per_second": 6.737, |
| "eval_math_steps_per_second": 1.684, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_physics_entropy": 0.8692562022209167, |
| "eval_physics_loss": 0.9228239059448242, |
| "eval_physics_mean_token_accuracy": 0.7882768630981445, |
| "eval_physics_num_tokens": 49671048.0, |
| "eval_physics_runtime": 85.556, |
| "eval_physics_samples_per_second": 5.844, |
| "eval_physics_steps_per_second": 1.461, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.8500566266477108, |
| "epoch": 2.416, |
| "grad_norm": 11.4375, |
| "learning_rate": 1.1044444444444444e-05, |
| "loss": 14.158, |
| "mean_token_accuracy": 0.7961105518043041, |
| "num_tokens": 50018182.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.8643727529793978, |
| "epoch": 2.432, |
| "grad_norm": 12.75, |
| "learning_rate": 1.0970370370370371e-05, |
| "loss": 14.4046, |
| "mean_token_accuracy": 0.7947129063308239, |
| "num_tokens": 50341187.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.8951457351446152, |
| "epoch": 2.448, |
| "grad_norm": 11.25, |
| "learning_rate": 1.0896296296296298e-05, |
| "loss": 14.8649, |
| "mean_token_accuracy": 0.7867488101124763, |
| "num_tokens": 50667019.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.8816273506730795, |
| "epoch": 2.464, |
| "grad_norm": 12.8125, |
| "learning_rate": 1.0822222222222222e-05, |
| "loss": 14.6, |
| "mean_token_accuracy": 0.7916153628379107, |
| "num_tokens": 50991113.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.8664202645421029, |
| "epoch": 2.48, |
| "grad_norm": 12.3125, |
| "learning_rate": 1.074814814814815e-05, |
| "loss": 14.5346, |
| "mean_token_accuracy": 0.7921106025576592, |
| "num_tokens": 51319624.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.8818996708840132, |
| "epoch": 2.496, |
| "grad_norm": 11.625, |
| "learning_rate": 1.0674074074074074e-05, |
| "loss": 14.6185, |
| "mean_token_accuracy": 0.7901674829423427, |
| "num_tokens": 51654633.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.8544634876772761, |
| "epoch": 2.512, |
| "grad_norm": 13.25, |
| "learning_rate": 1.0600000000000002e-05, |
| "loss": 14.2895, |
| "mean_token_accuracy": 0.7949651554226875, |
| "num_tokens": 51980051.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.8599010031670332, |
| "epoch": 2.528, |
| "grad_norm": 11.5625, |
| "learning_rate": 1.0525925925925927e-05, |
| "loss": 14.2789, |
| "mean_token_accuracy": 0.7945577628910542, |
| "num_tokens": 52316109.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.8746189616620541, |
| "epoch": 2.544, |
| "grad_norm": 13.6875, |
| "learning_rate": 1.0451851851851853e-05, |
| "loss": 14.6318, |
| "mean_token_accuracy": 0.7903719995170831, |
| "num_tokens": 52634528.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.8605435255914926, |
| "epoch": 2.56, |
| "grad_norm": 12.0, |
| "learning_rate": 1.0377777777777778e-05, |
| "loss": 14.3017, |
| "mean_token_accuracy": 0.7939822390675545, |
| "num_tokens": 52964035.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_biology_entropy": 2.439929902076721, |
| "eval_biology_loss": 2.855186939239502, |
| "eval_biology_mean_token_accuracy": 0.5345207722187042, |
| "eval_biology_num_tokens": 52964035.0, |
| "eval_biology_runtime": 57.4094, |
| "eval_biology_samples_per_second": 8.709, |
| "eval_biology_steps_per_second": 2.177, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_chemistry_entropy": 1.1931069812774657, |
| "eval_chemistry_loss": 1.278794527053833, |
| "eval_chemistry_mean_token_accuracy": 0.7340463919639587, |
| "eval_chemistry_num_tokens": 52964035.0, |
| "eval_chemistry_runtime": 71.3365, |
| "eval_chemistry_samples_per_second": 7.009, |
| "eval_chemistry_steps_per_second": 1.752, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_math_entropy": 0.8954266090393066, |
| "eval_math_loss": 1.1417946815490723, |
| "eval_math_mean_token_accuracy": 0.7577371735572815, |
| "eval_math_num_tokens": 52964035.0, |
| "eval_math_runtime": 74.3663, |
| "eval_math_samples_per_second": 6.723, |
| "eval_math_steps_per_second": 1.681, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_physics_entropy": 0.8706050214767456, |
| "eval_physics_loss": 0.9212888479232788, |
| "eval_physics_mean_token_accuracy": 0.7886158428192138, |
| "eval_physics_num_tokens": 52964035.0, |
| "eval_physics_runtime": 85.0105, |
| "eval_physics_samples_per_second": 5.882, |
| "eval_physics_steps_per_second": 1.47, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.8518974561244249, |
| "epoch": 2.576, |
| "grad_norm": 12.8125, |
| "learning_rate": 1.0303703703703705e-05, |
| "loss": 14.1697, |
| "mean_token_accuracy": 0.7962028808891773, |
| "num_tokens": 53293345.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.8667042430490255, |
| "epoch": 2.592, |
| "grad_norm": 10.625, |
| "learning_rate": 1.0229629629629631e-05, |
| "loss": 14.4307, |
| "mean_token_accuracy": 0.7936250735074282, |
| "num_tokens": 53620896.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.8402661351487041, |
| "epoch": 2.608, |
| "grad_norm": 11.25, |
| "learning_rate": 1.0155555555555556e-05, |
| "loss": 14.021, |
| "mean_token_accuracy": 0.7978927873075008, |
| "num_tokens": 53950989.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.8603313663974405, |
| "epoch": 2.624, |
| "grad_norm": 11.125, |
| "learning_rate": 1.0081481481481484e-05, |
| "loss": 14.3819, |
| "mean_token_accuracy": 0.7938243903219699, |
| "num_tokens": 54282732.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.8600944321602583, |
| "epoch": 2.64, |
| "grad_norm": 12.8125, |
| "learning_rate": 1.0007407407407407e-05, |
| "loss": 14.2866, |
| "mean_token_accuracy": 0.7947955295443535, |
| "num_tokens": 54609748.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.8892827957868576, |
| "epoch": 2.656, |
| "grad_norm": 12.5625, |
| "learning_rate": 9.933333333333334e-06, |
| "loss": 14.8315, |
| "mean_token_accuracy": 0.7884532894939185, |
| "num_tokens": 54953583.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.867825073376298, |
| "epoch": 2.672, |
| "grad_norm": 12.0, |
| "learning_rate": 9.85925925925926e-06, |
| "loss": 14.4804, |
| "mean_token_accuracy": 0.791333881765604, |
| "num_tokens": 55281945.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.8512159805744887, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 12.0625, |
| "learning_rate": 9.785185185185187e-06, |
| "loss": 14.0027, |
| "mean_token_accuracy": 0.798706853762269, |
| "num_tokens": 55612391.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.8470454571768642, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 11.25, |
| "learning_rate": 9.711111111111111e-06, |
| "loss": 14.2474, |
| "mean_token_accuracy": 0.7953305941075086, |
| "num_tokens": 55949791.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.8569748856127262, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 14.625, |
| "learning_rate": 9.637037037037038e-06, |
| "loss": 14.2518, |
| "mean_token_accuracy": 0.7947139970958232, |
| "num_tokens": 56282204.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_biology_entropy": 2.4623048429489134, |
| "eval_biology_loss": 2.855733633041382, |
| "eval_biology_mean_token_accuracy": 0.5338225281238556, |
| "eval_biology_num_tokens": 56282204.0, |
| "eval_biology_runtime": 57.2989, |
| "eval_biology_samples_per_second": 8.726, |
| "eval_biology_steps_per_second": 2.182, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_chemistry_entropy": 1.1981362085342406, |
| "eval_chemistry_loss": 1.2777812480926514, |
| "eval_chemistry_mean_token_accuracy": 0.7337445435523987, |
| "eval_chemistry_num_tokens": 56282204.0, |
| "eval_chemistry_runtime": 72.002, |
| "eval_chemistry_samples_per_second": 6.944, |
| "eval_chemistry_steps_per_second": 1.736, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_math_entropy": 0.8990244059562683, |
| "eval_math_loss": 1.1414705514907837, |
| "eval_math_mean_token_accuracy": 0.7577006416320801, |
| "eval_math_num_tokens": 56282204.0, |
| "eval_math_runtime": 74.1316, |
| "eval_math_samples_per_second": 6.745, |
| "eval_math_steps_per_second": 1.686, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_physics_entropy": 0.8712537698745727, |
| "eval_physics_loss": 0.9199900031089783, |
| "eval_physics_mean_token_accuracy": 0.7888414950370789, |
| "eval_physics_num_tokens": 56282204.0, |
| "eval_physics_runtime": 85.5277, |
| "eval_physics_samples_per_second": 5.846, |
| "eval_physics_steps_per_second": 1.462, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.8889217928051949, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 12.8125, |
| "learning_rate": 9.562962962962965e-06, |
| "loss": 14.8794, |
| "mean_token_accuracy": 0.787835606187582, |
| "num_tokens": 56605478.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.867193017527461, |
| "epoch": 2.752, |
| "grad_norm": 11.6875, |
| "learning_rate": 9.48888888888889e-06, |
| "loss": 14.5299, |
| "mean_token_accuracy": 0.7921088628470898, |
| "num_tokens": 56935874.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.8495670940726996, |
| "epoch": 2.768, |
| "grad_norm": 12.0, |
| "learning_rate": 9.414814814814816e-06, |
| "loss": 14.0654, |
| "mean_token_accuracy": 0.7975914411246776, |
| "num_tokens": 57270552.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.8686243133619428, |
| "epoch": 2.784, |
| "grad_norm": 12.5, |
| "learning_rate": 9.34074074074074e-06, |
| "loss": 14.4341, |
| "mean_token_accuracy": 0.7923915140330792, |
| "num_tokens": 57602968.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.9133183425292373, |
| "epoch": 2.8, |
| "grad_norm": 13.375, |
| "learning_rate": 9.266666666666667e-06, |
| "loss": 15.2497, |
| "mean_token_accuracy": 0.7833729684352875, |
| "num_tokens": 57925886.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.8448837269097567, |
| "epoch": 2.816, |
| "grad_norm": 11.875, |
| "learning_rate": 9.192592592592594e-06, |
| "loss": 13.9445, |
| "mean_token_accuracy": 0.7979842260479927, |
| "num_tokens": 58262960.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.871445388160646, |
| "epoch": 2.832, |
| "grad_norm": 12.3125, |
| "learning_rate": 9.118518518518518e-06, |
| "loss": 14.5598, |
| "mean_token_accuracy": 0.7907900519669055, |
| "num_tokens": 58586324.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.8576038857921958, |
| "epoch": 2.848, |
| "grad_norm": 13.0, |
| "learning_rate": 9.044444444444445e-06, |
| "loss": 14.3799, |
| "mean_token_accuracy": 0.7930273406207562, |
| "num_tokens": 58912949.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.8503576427698135, |
| "epoch": 2.864, |
| "grad_norm": 11.0, |
| "learning_rate": 8.970370370370372e-06, |
| "loss": 14.1151, |
| "mean_token_accuracy": 0.7967234510928393, |
| "num_tokens": 59252494.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.865095803141594, |
| "epoch": 2.88, |
| "grad_norm": 14.125, |
| "learning_rate": 8.896296296296298e-06, |
| "loss": 14.4, |
| "mean_token_accuracy": 0.7936870157718658, |
| "num_tokens": 59574951.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_biology_entropy": 2.4605735635757444, |
| "eval_biology_loss": 2.8438503742218018, |
| "eval_biology_mean_token_accuracy": 0.5349292945861817, |
| "eval_biology_num_tokens": 59574951.0, |
| "eval_biology_runtime": 57.1516, |
| "eval_biology_samples_per_second": 8.749, |
| "eval_biology_steps_per_second": 2.187, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_chemistry_entropy": 1.1904199600219727, |
| "eval_chemistry_loss": 1.2762060165405273, |
| "eval_chemistry_mean_token_accuracy": 0.7342206130027771, |
| "eval_chemistry_num_tokens": 59574951.0, |
| "eval_chemistry_runtime": 71.4215, |
| "eval_chemistry_samples_per_second": 7.001, |
| "eval_chemistry_steps_per_second": 1.75, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_math_entropy": 0.8948101491928101, |
| "eval_math_loss": 1.1407443284988403, |
| "eval_math_mean_token_accuracy": 0.7578761463165283, |
| "eval_math_num_tokens": 59574951.0, |
| "eval_math_runtime": 74.9351, |
| "eval_math_samples_per_second": 6.672, |
| "eval_math_steps_per_second": 1.668, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_physics_entropy": 0.8652872314453125, |
| "eval_physics_loss": 0.9192015528678894, |
| "eval_physics_mean_token_accuracy": 0.7890026559829711, |
| "eval_physics_num_tokens": 59574951.0, |
| "eval_physics_runtime": 84.6211, |
| "eval_physics_samples_per_second": 5.909, |
| "eval_physics_steps_per_second": 1.477, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.8504932520911097, |
| "epoch": 2.896, |
| "grad_norm": 12.75, |
| "learning_rate": 8.822222222222223e-06, |
| "loss": 14.2948, |
| "mean_token_accuracy": 0.7946070168167353, |
| "num_tokens": 59903274.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.8513661539182067, |
| "epoch": 2.912, |
| "grad_norm": 12.5625, |
| "learning_rate": 8.74814814814815e-06, |
| "loss": 14.1948, |
| "mean_token_accuracy": 0.7956477370113134, |
| "num_tokens": 60234203.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.8567061256617308, |
| "epoch": 2.928, |
| "grad_norm": 12.0625, |
| "learning_rate": 8.674074074074074e-06, |
| "loss": 14.2065, |
| "mean_token_accuracy": 0.7965482659637928, |
| "num_tokens": 60559123.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.8522207867354155, |
| "epoch": 2.944, |
| "grad_norm": 13.8125, |
| "learning_rate": 8.6e-06, |
| "loss": 14.1203, |
| "mean_token_accuracy": 0.7966358475387096, |
| "num_tokens": 60893423.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.8581688396632672, |
| "epoch": 2.96, |
| "grad_norm": 11.875, |
| "learning_rate": 8.525925925925927e-06, |
| "loss": 14.4079, |
| "mean_token_accuracy": 0.7932413015514612, |
| "num_tokens": 61228692.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.8690817683935166, |
| "epoch": 2.976, |
| "grad_norm": 11.9375, |
| "learning_rate": 8.451851851851852e-06, |
| "loss": 14.3962, |
| "mean_token_accuracy": 0.793168543279171, |
| "num_tokens": 61560852.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.8346356390044093, |
| "epoch": 2.992, |
| "grad_norm": 11.9375, |
| "learning_rate": 8.377777777777779e-06, |
| "loss": 13.939, |
| "mean_token_accuracy": 0.7977076359093189, |
| "num_tokens": 61898951.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.8470860140398144, |
| "epoch": 3.008, |
| "grad_norm": 11.4375, |
| "learning_rate": 8.303703703703705e-06, |
| "loss": 14.0544, |
| "mean_token_accuracy": 0.7982985734939575, |
| "num_tokens": 62223875.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.8579535936936736, |
| "epoch": 3.024, |
| "grad_norm": 12.0625, |
| "learning_rate": 8.229629629629632e-06, |
| "loss": 14.2346, |
| "mean_token_accuracy": 0.7953595809638501, |
| "num_tokens": 62553166.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.8405753003433347, |
| "epoch": 3.04, |
| "grad_norm": 12.0625, |
| "learning_rate": 8.155555555555556e-06, |
| "loss": 13.9614, |
| "mean_token_accuracy": 0.7975836973637342, |
| "num_tokens": 62893113.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_biology_entropy": 2.466928297996521, |
| "eval_biology_loss": 2.8451387882232666, |
| "eval_biology_mean_token_accuracy": 0.5350441763401032, |
| "eval_biology_num_tokens": 62893113.0, |
| "eval_biology_runtime": 57.2798, |
| "eval_biology_samples_per_second": 8.729, |
| "eval_biology_steps_per_second": 2.182, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_chemistry_entropy": 1.1924911136627196, |
| "eval_chemistry_loss": 1.2755722999572754, |
| "eval_chemistry_mean_token_accuracy": 0.734372947216034, |
| "eval_chemistry_num_tokens": 62893113.0, |
| "eval_chemistry_runtime": 71.12, |
| "eval_chemistry_samples_per_second": 7.03, |
| "eval_chemistry_steps_per_second": 1.758, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_math_entropy": 0.8945438508987427, |
| "eval_math_loss": 1.1409516334533691, |
| "eval_math_mean_token_accuracy": 0.7579452800750732, |
| "eval_math_num_tokens": 62893113.0, |
| "eval_math_runtime": 74.6477, |
| "eval_math_samples_per_second": 6.698, |
| "eval_math_steps_per_second": 1.675, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_physics_entropy": 0.8654837455749512, |
| "eval_physics_loss": 0.9181889295578003, |
| "eval_physics_mean_token_accuracy": 0.7893532915115357, |
| "eval_physics_num_tokens": 62893113.0, |
| "eval_physics_runtime": 85.3356, |
| "eval_physics_samples_per_second": 5.859, |
| "eval_physics_steps_per_second": 1.465, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.8652172423899174, |
| "epoch": 3.056, |
| "grad_norm": 12.375, |
| "learning_rate": 8.081481481481483e-06, |
| "loss": 14.4707, |
| "mean_token_accuracy": 0.7926124636083841, |
| "num_tokens": 63218088.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.8625176247209311, |
| "epoch": 3.072, |
| "grad_norm": 13.5, |
| "learning_rate": 8.007407407407408e-06, |
| "loss": 14.2819, |
| "mean_token_accuracy": 0.7927148371934891, |
| "num_tokens": 63551201.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.8709696922451258, |
| "epoch": 3.088, |
| "grad_norm": 13.25, |
| "learning_rate": 7.933333333333334e-06, |
| "loss": 14.535, |
| "mean_token_accuracy": 0.7909463763236999, |
| "num_tokens": 63876362.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.8365857848897577, |
| "epoch": 3.104, |
| "grad_norm": 12.6875, |
| "learning_rate": 7.859259259259259e-06, |
| "loss": 13.917, |
| "mean_token_accuracy": 0.7985875517129898, |
| "num_tokens": 64208171.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.8436792023479939, |
| "epoch": 3.12, |
| "grad_norm": 12.5, |
| "learning_rate": 7.785185185185185e-06, |
| "loss": 13.9247, |
| "mean_token_accuracy": 0.7994871154427529, |
| "num_tokens": 64544493.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.8724946033209562, |
| "epoch": 3.136, |
| "grad_norm": 12.875, |
| "learning_rate": 7.711111111111112e-06, |
| "loss": 14.5896, |
| "mean_token_accuracy": 0.790136469900608, |
| "num_tokens": 64872488.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.8340439153835177, |
| "epoch": 3.152, |
| "grad_norm": 12.6875, |
| "learning_rate": 7.637037037037037e-06, |
| "loss": 13.9309, |
| "mean_token_accuracy": 0.7989891562610865, |
| "num_tokens": 65201124.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.8577351365238428, |
| "epoch": 3.168, |
| "grad_norm": 11.125, |
| "learning_rate": 7.562962962962963e-06, |
| "loss": 14.332, |
| "mean_token_accuracy": 0.7956987973302603, |
| "num_tokens": 65525868.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.8472510496154427, |
| "epoch": 3.184, |
| "grad_norm": 12.75, |
| "learning_rate": 7.48888888888889e-06, |
| "loss": 14.0577, |
| "mean_token_accuracy": 0.7991137109696865, |
| "num_tokens": 65861603.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.8585745912045241, |
| "epoch": 3.2, |
| "grad_norm": 11.3125, |
| "learning_rate": 7.4148148148148155e-06, |
| "loss": 14.2601, |
| "mean_token_accuracy": 0.7946750316768885, |
| "num_tokens": 66193675.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_biology_entropy": 2.4529026355743406, |
| "eval_biology_loss": 2.8515477180480957, |
| "eval_biology_mean_token_accuracy": 0.5361672310829163, |
| "eval_biology_num_tokens": 66193675.0, |
| "eval_biology_runtime": 56.9907, |
| "eval_biology_samples_per_second": 8.773, |
| "eval_biology_steps_per_second": 2.193, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_chemistry_entropy": 1.1896215171813964, |
| "eval_chemistry_loss": 1.2748806476593018, |
| "eval_chemistry_mean_token_accuracy": 0.7352399125099182, |
| "eval_chemistry_num_tokens": 66193675.0, |
| "eval_chemistry_runtime": 71.1598, |
| "eval_chemistry_samples_per_second": 7.026, |
| "eval_chemistry_steps_per_second": 1.757, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_math_entropy": 0.8935150079727173, |
| "eval_math_loss": 1.140594482421875, |
| "eval_math_mean_token_accuracy": 0.7576922287940979, |
| "eval_math_num_tokens": 66193675.0, |
| "eval_math_runtime": 74.3826, |
| "eval_math_samples_per_second": 6.722, |
| "eval_math_steps_per_second": 1.681, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_physics_entropy": 0.8648786902427673, |
| "eval_physics_loss": 0.9176769852638245, |
| "eval_physics_mean_token_accuracy": 0.7893374056816101, |
| "eval_physics_num_tokens": 66193675.0, |
| "eval_physics_runtime": 84.6448, |
| "eval_physics_samples_per_second": 5.907, |
| "eval_physics_steps_per_second": 1.477, |
| "step": 2000 |
| } |
| ], |
| "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": 6.614072648666411e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
