| { |
| "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": 1.3688963137567043, |
| "epoch": 0.016, |
| "grad_norm": 26.75, |
| "learning_rate": 1.8e-07, |
| "loss": 2.1543, |
| "mean_token_accuracy": 0.6007782766595483, |
| "num_tokens": 195510.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.3619943611323833, |
| "epoch": 0.032, |
| "grad_norm": 22.625, |
| "learning_rate": 3.8e-07, |
| "loss": 2.1586, |
| "mean_token_accuracy": 0.6013051267713309, |
| "num_tokens": 391008.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.4113613955676556, |
| "epoch": 0.048, |
| "grad_norm": 20.875, |
| "learning_rate": 5.800000000000001e-07, |
| "loss": 2.132, |
| "mean_token_accuracy": 0.6053086685016751, |
| "num_tokens": 590136.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.4471838753670454, |
| "epoch": 0.064, |
| "grad_norm": 17.125, |
| "learning_rate": 7.8e-07, |
| "loss": 2.0923, |
| "mean_token_accuracy": 0.6051462920382619, |
| "num_tokens": 791432.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.5650674737989902, |
| "epoch": 0.08, |
| "grad_norm": 13.625, |
| "learning_rate": 9.800000000000001e-07, |
| "loss": 2.0495, |
| "mean_token_accuracy": 0.6057845208793878, |
| "num_tokens": 990342.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.7182099640369415, |
| "epoch": 0.096, |
| "grad_norm": 8.8125, |
| "learning_rate": 1.1800000000000001e-06, |
| "loss": 2.0641, |
| "mean_token_accuracy": 0.604715196043253, |
| "num_tokens": 1182339.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.8127950172871352, |
| "epoch": 0.112, |
| "grad_norm": 5.875, |
| "learning_rate": 1.3800000000000001e-06, |
| "loss": 1.959, |
| "mean_token_accuracy": 0.6124658133834601, |
| "num_tokens": 1386193.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.8337691940367222, |
| "epoch": 0.128, |
| "grad_norm": 5.8125, |
| "learning_rate": 1.5800000000000001e-06, |
| "loss": 1.9629, |
| "mean_token_accuracy": 0.610929672792554, |
| "num_tokens": 1583261.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.79930883012712, |
| "epoch": 0.144, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.7800000000000001e-06, |
| "loss": 1.9463, |
| "mean_token_accuracy": 0.6148752361536026, |
| "num_tokens": 1774826.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.8165954321622848, |
| "epoch": 0.16, |
| "grad_norm": 5.90625, |
| "learning_rate": 1.98e-06, |
| "loss": 1.9617, |
| "mean_token_accuracy": 0.6118428658694028, |
| "num_tokens": 1971310.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 1.7495100326538087, |
| "eval_biology_loss": 1.9257053136825562, |
| "eval_biology_mean_token_accuracy": 0.6249015061855316, |
| "eval_biology_num_tokens": 1971310.0, |
| "eval_biology_runtime": 39.1194, |
| "eval_biology_samples_per_second": 12.781, |
| "eval_biology_steps_per_second": 3.195, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 0.7336231064796448, |
| "eval_chemistry_loss": 0.8070052266120911, |
| "eval_chemistry_mean_token_accuracy": 0.8109221415519714, |
| "eval_chemistry_num_tokens": 1971310.0, |
| "eval_chemistry_runtime": 48.7021, |
| "eval_chemistry_samples_per_second": 10.266, |
| "eval_chemistry_steps_per_second": 2.567, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.7839918725192547, |
| "epoch": 0.176, |
| "grad_norm": 4.75, |
| "learning_rate": 2.1800000000000003e-06, |
| "loss": 1.911, |
| "mean_token_accuracy": 0.6167426453903317, |
| "num_tokens": 2169453.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 1.798803521692753, |
| "epoch": 0.192, |
| "grad_norm": 5.59375, |
| "learning_rate": 2.38e-06, |
| "loss": 1.9179, |
| "mean_token_accuracy": 0.6152951141819358, |
| "num_tokens": 2366844.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.7993743151426316, |
| "epoch": 0.208, |
| "grad_norm": 5.21875, |
| "learning_rate": 2.5800000000000003e-06, |
| "loss": 1.9124, |
| "mean_token_accuracy": 0.6123070493340492, |
| "num_tokens": 2559708.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.7579403154551982, |
| "epoch": 0.224, |
| "grad_norm": 5.5, |
| "learning_rate": 2.7800000000000005e-06, |
| "loss": 1.8823, |
| "mean_token_accuracy": 0.6201574027538299, |
| "num_tokens": 2756533.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.751310784369707, |
| "epoch": 0.24, |
| "grad_norm": 5.0, |
| "learning_rate": 2.9800000000000003e-06, |
| "loss": 1.8497, |
| "mean_token_accuracy": 0.6232937585562468, |
| "num_tokens": 2948600.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.7673518907278776, |
| "epoch": 0.256, |
| "grad_norm": 5.21875, |
| "learning_rate": 3.1800000000000005e-06, |
| "loss": 1.8636, |
| "mean_token_accuracy": 0.6187300987541675, |
| "num_tokens": 3141473.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.7522739164531231, |
| "epoch": 0.272, |
| "grad_norm": 5.0, |
| "learning_rate": 3.3800000000000007e-06, |
| "loss": 1.8404, |
| "mean_token_accuracy": 0.6195556577295065, |
| "num_tokens": 3337285.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.735862059891224, |
| "epoch": 0.288, |
| "grad_norm": 4.9375, |
| "learning_rate": 3.58e-06, |
| "loss": 1.8084, |
| "mean_token_accuracy": 0.6230981133878231, |
| "num_tokens": 3541221.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.7118580989539622, |
| "epoch": 0.304, |
| "grad_norm": 4.65625, |
| "learning_rate": 3.7800000000000002e-06, |
| "loss": 1.7784, |
| "mean_token_accuracy": 0.626662303134799, |
| "num_tokens": 3734952.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.7322832621634006, |
| "epoch": 0.32, |
| "grad_norm": 4.78125, |
| "learning_rate": 3.980000000000001e-06, |
| "loss": 1.7879, |
| "mean_token_accuracy": 0.623782691732049, |
| "num_tokens": 3922209.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 1.703305299282074, |
| "eval_biology_loss": 1.7626646757125854, |
| "eval_biology_mean_token_accuracy": 0.635826956987381, |
| "eval_biology_num_tokens": 3922209.0, |
| "eval_biology_runtime": 39.0903, |
| "eval_biology_samples_per_second": 12.791, |
| "eval_biology_steps_per_second": 3.198, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 0.7484376595020295, |
| "eval_chemistry_loss": 0.7622011303901672, |
| "eval_chemistry_mean_token_accuracy": 0.8104680519104004, |
| "eval_chemistry_num_tokens": 3922209.0, |
| "eval_chemistry_runtime": 48.6797, |
| "eval_chemistry_samples_per_second": 10.271, |
| "eval_chemistry_steps_per_second": 2.568, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.6908923149108888, |
| "epoch": 0.336, |
| "grad_norm": 4.6875, |
| "learning_rate": 4.18e-06, |
| "loss": 1.7343, |
| "mean_token_accuracy": 0.6304708372801542, |
| "num_tokens": 4115941.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.7183935269713402, |
| "epoch": 0.352, |
| "grad_norm": 5.0, |
| "learning_rate": 4.38e-06, |
| "loss": 1.7323, |
| "mean_token_accuracy": 0.6271383590996266, |
| "num_tokens": 4308755.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 1.7019562765955925, |
| "epoch": 0.368, |
| "grad_norm": 4.15625, |
| "learning_rate": 4.58e-06, |
| "loss": 1.7129, |
| "mean_token_accuracy": 0.6312347020953893, |
| "num_tokens": 4505910.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.699073526263237, |
| "epoch": 0.384, |
| "grad_norm": 4.78125, |
| "learning_rate": 4.78e-06, |
| "loss": 1.6857, |
| "mean_token_accuracy": 0.6311217818409205, |
| "num_tokens": 4696101.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.7013882726430893, |
| "epoch": 0.4, |
| "grad_norm": 4.375, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 1.677, |
| "mean_token_accuracy": 0.6308177687227726, |
| "num_tokens": 4889445.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 1.7109287835657596, |
| "epoch": 0.416, |
| "grad_norm": 4.28125, |
| "learning_rate": 5.18e-06, |
| "loss": 1.6829, |
| "mean_token_accuracy": 0.6278879173099995, |
| "num_tokens": 5087624.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 1.7145581994205714, |
| "epoch": 0.432, |
| "grad_norm": 5.0625, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 1.681, |
| "mean_token_accuracy": 0.6289313135668635, |
| "num_tokens": 5273794.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 1.7117976807057858, |
| "epoch": 0.448, |
| "grad_norm": 5.15625, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 1.6672, |
| "mean_token_accuracy": 0.6299384724348783, |
| "num_tokens": 5463288.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 1.642435162886977, |
| "epoch": 0.464, |
| "grad_norm": 4.4375, |
| "learning_rate": 5.78e-06, |
| "loss": 1.6048, |
| "mean_token_accuracy": 0.641257194057107, |
| "num_tokens": 5656827.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 1.6684199325740336, |
| "epoch": 0.48, |
| "grad_norm": 4.3125, |
| "learning_rate": 5.98e-06, |
| "loss": 1.6272, |
| "mean_token_accuracy": 0.6384812213480473, |
| "num_tokens": 5853311.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 1.6503952503204347, |
| "eval_biology_loss": 1.6127734184265137, |
| "eval_biology_mean_token_accuracy": 0.6489202432632446, |
| "eval_biology_num_tokens": 5853311.0, |
| "eval_biology_runtime": 39.0858, |
| "eval_biology_samples_per_second": 12.792, |
| "eval_biology_steps_per_second": 3.198, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 0.7682121543884277, |
| "eval_chemistry_loss": 0.7135673761367798, |
| "eval_chemistry_mean_token_accuracy": 0.8120616941452027, |
| "eval_chemistry_num_tokens": 5853311.0, |
| "eval_chemistry_runtime": 48.6951, |
| "eval_chemistry_samples_per_second": 10.268, |
| "eval_chemistry_steps_per_second": 2.567, |
| "step": 300 |
| }, |
| { |
| "entropy": 1.5807546474039555, |
| "epoch": 0.496, |
| "grad_norm": 4.6875, |
| "learning_rate": 6.18e-06, |
| "loss": 1.5368, |
| "mean_token_accuracy": 0.6513836592435837, |
| "num_tokens": 6049788.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 1.678506825864315, |
| "epoch": 0.512, |
| "grad_norm": 4.65625, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 1.6246, |
| "mean_token_accuracy": 0.6385043509304523, |
| "num_tokens": 6243825.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 1.652913999557495, |
| "epoch": 0.528, |
| "grad_norm": 4.4375, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 1.6009, |
| "mean_token_accuracy": 0.64362001568079, |
| "num_tokens": 6434188.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 1.6507954083383083, |
| "epoch": 0.544, |
| "grad_norm": 4.25, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 1.6102, |
| "mean_token_accuracy": 0.6433101873844862, |
| "num_tokens": 6629832.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 1.6201883230358363, |
| "epoch": 0.56, |
| "grad_norm": 4.28125, |
| "learning_rate": 6.98e-06, |
| "loss": 1.5677, |
| "mean_token_accuracy": 0.64635454043746, |
| "num_tokens": 6824668.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 1.589719269424677, |
| "epoch": 0.576, |
| "grad_norm": 4.375, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 1.5451, |
| "mean_token_accuracy": 0.6491251610219478, |
| "num_tokens": 7025923.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 1.6323727253824472, |
| "epoch": 0.592, |
| "grad_norm": 4.0625, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 1.5618, |
| "mean_token_accuracy": 0.645591801777482, |
| "num_tokens": 7218224.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 1.6019253939390183, |
| "epoch": 0.608, |
| "grad_norm": 4.65625, |
| "learning_rate": 7.58e-06, |
| "loss": 1.562, |
| "mean_token_accuracy": 0.6492579754441976, |
| "num_tokens": 7421216.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 1.6110554873943328, |
| "epoch": 0.624, |
| "grad_norm": 4.375, |
| "learning_rate": 7.78e-06, |
| "loss": 1.5511, |
| "mean_token_accuracy": 0.6453482262790203, |
| "num_tokens": 7617258.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 1.6183445014059543, |
| "epoch": 0.64, |
| "grad_norm": 4.40625, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 1.5462, |
| "mean_token_accuracy": 0.6479769114404916, |
| "num_tokens": 7806173.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 1.5254363222122191, |
| "eval_biology_loss": 1.5303529500961304, |
| "eval_biology_mean_token_accuracy": 0.6589018092155456, |
| "eval_biology_num_tokens": 7806173.0, |
| "eval_biology_runtime": 39.136, |
| "eval_biology_samples_per_second": 12.776, |
| "eval_biology_steps_per_second": 3.194, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 0.7431424338817596, |
| "eval_chemistry_loss": 0.7057921290397644, |
| "eval_chemistry_mean_token_accuracy": 0.8118511362075805, |
| "eval_chemistry_num_tokens": 7806173.0, |
| "eval_chemistry_runtime": 48.743, |
| "eval_chemistry_samples_per_second": 10.258, |
| "eval_chemistry_steps_per_second": 2.564, |
| "step": 400 |
| }, |
| { |
| "entropy": 1.5993809100240468, |
| "epoch": 0.656, |
| "grad_norm": 4.6875, |
| "learning_rate": 8.18e-06, |
| "loss": 1.5514, |
| "mean_token_accuracy": 0.6487635668367148, |
| "num_tokens": 8000514.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 1.6190266445279122, |
| "epoch": 0.672, |
| "grad_norm": 4.40625, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 1.5587, |
| "mean_token_accuracy": 0.6449675388634205, |
| "num_tokens": 8187884.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 1.5749193426221608, |
| "epoch": 0.688, |
| "grad_norm": 4.125, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 1.523, |
| "mean_token_accuracy": 0.651202504709363, |
| "num_tokens": 8390643.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 1.5851164262741804, |
| "epoch": 0.704, |
| "grad_norm": 4.375, |
| "learning_rate": 8.78e-06, |
| "loss": 1.5246, |
| "mean_token_accuracy": 0.6505646403878927, |
| "num_tokens": 8583266.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 1.500511408224702, |
| "epoch": 0.72, |
| "grad_norm": 3.875, |
| "learning_rate": 8.98e-06, |
| "loss": 1.4375, |
| "mean_token_accuracy": 0.664596076309681, |
| "num_tokens": 8786233.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 1.5476290859282016, |
| "epoch": 0.736, |
| "grad_norm": 4.5625, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 1.476, |
| "mean_token_accuracy": 0.6562888745218516, |
| "num_tokens": 8982898.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 1.5444358900189399, |
| "epoch": 0.752, |
| "grad_norm": 4.625, |
| "learning_rate": 9.38e-06, |
| "loss": 1.4792, |
| "mean_token_accuracy": 0.6567551180720329, |
| "num_tokens": 9174361.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 1.494469254463911, |
| "epoch": 0.768, |
| "grad_norm": 4.125, |
| "learning_rate": 9.58e-06, |
| "loss": 1.4385, |
| "mean_token_accuracy": 0.6629225756973028, |
| "num_tokens": 9373430.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 1.5377752494066954, |
| "epoch": 0.784, |
| "grad_norm": 4.21875, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 1.4748, |
| "mean_token_accuracy": 0.6564478818327188, |
| "num_tokens": 9570409.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 1.546391127258539, |
| "epoch": 0.8, |
| "grad_norm": 4.46875, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 1.4842, |
| "mean_token_accuracy": 0.6546981681138277, |
| "num_tokens": 9766360.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 1.4740597500801087, |
| "eval_biology_loss": 1.4621686935424805, |
| "eval_biology_mean_token_accuracy": 0.6668880858421325, |
| "eval_biology_num_tokens": 9766360.0, |
| "eval_biology_runtime": 39.1212, |
| "eval_biology_samples_per_second": 12.781, |
| "eval_biology_steps_per_second": 3.195, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 0.739832299232483, |
| "eval_chemistry_loss": 0.7017239928245544, |
| "eval_chemistry_mean_token_accuracy": 0.8113184819221496, |
| "eval_chemistry_num_tokens": 9766360.0, |
| "eval_chemistry_runtime": 48.5413, |
| "eval_chemistry_samples_per_second": 10.301, |
| "eval_chemistry_steps_per_second": 2.575, |
| "step": 500 |
| }, |
| { |
| "entropy": 1.4985189002007246, |
| "epoch": 0.816, |
| "grad_norm": 4.625, |
| "learning_rate": 1.018e-05, |
| "loss": 1.4466, |
| "mean_token_accuracy": 0.6619349274784326, |
| "num_tokens": 9963834.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 1.48775667399168, |
| "epoch": 0.832, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.038e-05, |
| "loss": 1.4375, |
| "mean_token_accuracy": 0.6627471141517163, |
| "num_tokens": 10160950.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 1.479408584907651, |
| "epoch": 0.848, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 1.4177, |
| "mean_token_accuracy": 0.6654143054038286, |
| "num_tokens": 10363131.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 1.5001295048743486, |
| "epoch": 0.864, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 1.4434, |
| "mean_token_accuracy": 0.6616969238966703, |
| "num_tokens": 10558003.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 1.4739857867360115, |
| "epoch": 0.88, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 1.4201, |
| "mean_token_accuracy": 0.665744898840785, |
| "num_tokens": 10754470.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 1.5231038693338632, |
| "epoch": 0.896, |
| "grad_norm": 4.5, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 1.4565, |
| "mean_token_accuracy": 0.6590576630085707, |
| "num_tokens": 10948943.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 1.4710557978600263, |
| "epoch": 0.912, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.138e-05, |
| "loss": 1.4183, |
| "mean_token_accuracy": 0.6641684602946043, |
| "num_tokens": 11142613.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 1.4895741816610097, |
| "epoch": 0.928, |
| "grad_norm": 3.984375, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 1.4255, |
| "mean_token_accuracy": 0.6618844125419855, |
| "num_tokens": 11336779.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 1.4350463204085826, |
| "epoch": 0.944, |
| "grad_norm": 3.96875, |
| "learning_rate": 1.178e-05, |
| "loss": 1.3876, |
| "mean_token_accuracy": 0.6704638510942459, |
| "num_tokens": 11536381.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 1.4527774393558501, |
| "epoch": 0.96, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.198e-05, |
| "loss": 1.401, |
| "mean_token_accuracy": 0.6658082317560912, |
| "num_tokens": 11735527.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 1.4157320256233215, |
| "eval_biology_loss": 1.4121752977371216, |
| "eval_biology_mean_token_accuracy": 0.6741212258338928, |
| "eval_biology_num_tokens": 11735527.0, |
| "eval_biology_runtime": 39.1158, |
| "eval_biology_samples_per_second": 12.783, |
| "eval_biology_steps_per_second": 3.196, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_chemistry_entropy": 0.7435694184303284, |
| "eval_chemistry_loss": 0.7019251585006714, |
| "eval_chemistry_mean_token_accuracy": 0.8104675717353821, |
| "eval_chemistry_num_tokens": 11735527.0, |
| "eval_chemistry_runtime": 48.7451, |
| "eval_chemistry_samples_per_second": 10.257, |
| "eval_chemistry_steps_per_second": 2.564, |
| "step": 600 |
| }, |
| { |
| "entropy": 1.4274786025285722, |
| "epoch": 0.976, |
| "grad_norm": 4.25, |
| "learning_rate": 1.218e-05, |
| "loss": 1.3729, |
| "mean_token_accuracy": 0.6719065200537443, |
| "num_tokens": 11930935.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 1.4573175739496946, |
| "epoch": 0.992, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 1.4041, |
| "mean_token_accuracy": 0.6670554719865323, |
| "num_tokens": 12129585.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 1.4293555557727813, |
| "epoch": 1.008, |
| "grad_norm": 3.796875, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 1.3792, |
| "mean_token_accuracy": 0.6710654832422733, |
| "num_tokens": 12325847.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 1.3706564407795667, |
| "epoch": 1.024, |
| "grad_norm": 3.96875, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 1.3138, |
| "mean_token_accuracy": 0.6818066164851189, |
| "num_tokens": 12530820.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 1.4076907094568014, |
| "epoch": 1.04, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 1.3577, |
| "mean_token_accuracy": 0.6723817244172097, |
| "num_tokens": 12725384.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 1.3665749713778497, |
| "epoch": 1.056, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 1.3016, |
| "mean_token_accuracy": 0.6819081131368876, |
| "num_tokens": 12924592.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 1.3977813702076674, |
| "epoch": 1.072, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 1.3368, |
| "mean_token_accuracy": 0.6750961408019066, |
| "num_tokens": 13112848.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 1.3841964561492204, |
| "epoch": 1.088, |
| "grad_norm": 3.859375, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 1.3301, |
| "mean_token_accuracy": 0.6785592496395111, |
| "num_tokens": 13305725.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 1.3711885944008828, |
| "epoch": 1.104, |
| "grad_norm": 4.0, |
| "learning_rate": 1.378e-05, |
| "loss": 1.3226, |
| "mean_token_accuracy": 0.6788452990353108, |
| "num_tokens": 13498685.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 1.3965423095971345, |
| "epoch": 1.12, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.398e-05, |
| "loss": 1.3255, |
| "mean_token_accuracy": 0.6777807965874671, |
| "num_tokens": 13682146.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_biology_entropy": 1.3078518261909484, |
| "eval_biology_loss": 1.3719801902770996, |
| "eval_biology_mean_token_accuracy": 0.6799422569274902, |
| "eval_biology_num_tokens": 13682146.0, |
| "eval_biology_runtime": 39.0883, |
| "eval_biology_samples_per_second": 12.792, |
| "eval_biology_steps_per_second": 3.198, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_chemistry_entropy": 0.6988301925659179, |
| "eval_chemistry_loss": 0.7011456489562988, |
| "eval_chemistry_mean_token_accuracy": 0.8096043920516968, |
| "eval_chemistry_num_tokens": 13682146.0, |
| "eval_chemistry_runtime": 48.7053, |
| "eval_chemistry_samples_per_second": 10.266, |
| "eval_chemistry_steps_per_second": 2.566, |
| "step": 700 |
| }, |
| { |
| "entropy": 1.3658908106386662, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.418e-05, |
| "loss": 1.322, |
| "mean_token_accuracy": 0.6789216380566359, |
| "num_tokens": 13876656.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 1.3465955082327128, |
| "epoch": 1.152, |
| "grad_norm": 3.890625, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 1.2917, |
| "mean_token_accuracy": 0.682017782703042, |
| "num_tokens": 14086104.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 1.3684371966868638, |
| "epoch": 1.168, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 1.314, |
| "mean_token_accuracy": 0.6794850911945105, |
| "num_tokens": 14274520.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 1.3601514294743537, |
| "epoch": 1.184, |
| "grad_norm": 3.65625, |
| "learning_rate": 1.478e-05, |
| "loss": 1.3094, |
| "mean_token_accuracy": 0.6805991049855947, |
| "num_tokens": 14473623.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 1.3518083952367306, |
| "epoch": 1.2, |
| "grad_norm": 3.828125, |
| "learning_rate": 1.498e-05, |
| "loss": 1.2965, |
| "mean_token_accuracy": 0.6813483159989119, |
| "num_tokens": 14661238.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 1.3173013672232627, |
| "epoch": 1.216, |
| "grad_norm": 3.796875, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 1.2626, |
| "mean_token_accuracy": 0.6857349291443825, |
| "num_tokens": 14865915.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 1.3409026943147182, |
| "epoch": 1.232, |
| "grad_norm": 3.625, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 1.2865, |
| "mean_token_accuracy": 0.6858917985111475, |
| "num_tokens": 15055682.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 1.344399669766426, |
| "epoch": 1.248, |
| "grad_norm": 3.875, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 1.2956, |
| "mean_token_accuracy": 0.6822925679385662, |
| "num_tokens": 15249576.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 1.3598602082580329, |
| "epoch": 1.264, |
| "grad_norm": 3.359375, |
| "learning_rate": 1.578e-05, |
| "loss": 1.2951, |
| "mean_token_accuracy": 0.6818954084068537, |
| "num_tokens": 15446149.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 1.36542206145823, |
| "epoch": 1.28, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 1.32, |
| "mean_token_accuracy": 0.6771101258695126, |
| "num_tokens": 15639317.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_biology_entropy": 1.3049154906272888, |
| "eval_biology_loss": 1.33707857131958, |
| "eval_biology_mean_token_accuracy": 0.683680823802948, |
| "eval_biology_num_tokens": 15639317.0, |
| "eval_biology_runtime": 39.1846, |
| "eval_biology_samples_per_second": 12.76, |
| "eval_biology_steps_per_second": 3.19, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_chemistry_entropy": 0.715079505443573, |
| "eval_chemistry_loss": 0.7021569609642029, |
| "eval_chemistry_mean_token_accuracy": 0.8091985578536988, |
| "eval_chemistry_num_tokens": 15639317.0, |
| "eval_chemistry_runtime": 48.799, |
| "eval_chemistry_samples_per_second": 10.246, |
| "eval_chemistry_steps_per_second": 2.562, |
| "step": 800 |
| }, |
| { |
| "entropy": 1.3627730403095484, |
| "epoch": 1.296, |
| "grad_norm": 3.375, |
| "learning_rate": 1.618e-05, |
| "loss": 1.3074, |
| "mean_token_accuracy": 0.6805873740464449, |
| "num_tokens": 15835805.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 1.3559819746762514, |
| "epoch": 1.312, |
| "grad_norm": 3.546875, |
| "learning_rate": 1.638e-05, |
| "loss": 1.2937, |
| "mean_token_accuracy": 0.6807271435856819, |
| "num_tokens": 16028375.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 1.356961441040039, |
| "epoch": 1.328, |
| "grad_norm": 3.546875, |
| "learning_rate": 1.658e-05, |
| "loss": 1.2983, |
| "mean_token_accuracy": 0.6812620472162962, |
| "num_tokens": 16230555.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 1.3171780712902545, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 3.390625, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 1.2739, |
| "mean_token_accuracy": 0.6877064649015665, |
| "num_tokens": 16436297.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 1.3393035437911749, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 3.90625, |
| "learning_rate": 1.698e-05, |
| "loss": 1.2744, |
| "mean_token_accuracy": 0.6847097586840392, |
| "num_tokens": 16630563.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 1.3660046834498645, |
| "epoch": 1.376, |
| "grad_norm": 3.859375, |
| "learning_rate": 1.718e-05, |
| "loss": 1.3048, |
| "mean_token_accuracy": 0.679730985313654, |
| "num_tokens": 16822602.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 1.3303084772080183, |
| "epoch": 1.392, |
| "grad_norm": 3.390625, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 1.2773, |
| "mean_token_accuracy": 0.6869184520095587, |
| "num_tokens": 17015750.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 1.3214296508580445, |
| "epoch": 1.408, |
| "grad_norm": 3.015625, |
| "learning_rate": 1.758e-05, |
| "loss": 1.2688, |
| "mean_token_accuracy": 0.6854220055043697, |
| "num_tokens": 17207248.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 1.3336139082908631, |
| "epoch": 1.424, |
| "grad_norm": 3.546875, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 1.29, |
| "mean_token_accuracy": 0.6841222684830427, |
| "num_tokens": 17403869.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 1.328351939469576, |
| "epoch": 1.44, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.798e-05, |
| "loss": 1.2696, |
| "mean_token_accuracy": 0.683881213143468, |
| "num_tokens": 17597428.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_biology_entropy": 1.293550329685211, |
| "eval_biology_loss": 1.3036084175109863, |
| "eval_biology_mean_token_accuracy": 0.6888437666893006, |
| "eval_biology_num_tokens": 17597428.0, |
| "eval_biology_runtime": 39.2009, |
| "eval_biology_samples_per_second": 12.755, |
| "eval_biology_steps_per_second": 3.189, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_chemistry_entropy": 0.711592486858368, |
| "eval_chemistry_loss": 0.7021830081939697, |
| "eval_chemistry_mean_token_accuracy": 0.8084347972869873, |
| "eval_chemistry_num_tokens": 17597428.0, |
| "eval_chemistry_runtime": 48.8241, |
| "eval_chemistry_samples_per_second": 10.241, |
| "eval_chemistry_steps_per_second": 2.56, |
| "step": 900 |
| }, |
| { |
| "entropy": 1.3218930535018445, |
| "epoch": 1.456, |
| "grad_norm": 3.515625, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 1.2763, |
| "mean_token_accuracy": 0.6863383412361145, |
| "num_tokens": 17798083.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 1.3118344832211732, |
| "epoch": 1.472, |
| "grad_norm": 3.265625, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 1.254, |
| "mean_token_accuracy": 0.6886441022157669, |
| "num_tokens": 17993802.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 1.309800398349762, |
| "epoch": 1.488, |
| "grad_norm": 3.453125, |
| "learning_rate": 1.858e-05, |
| "loss": 1.2575, |
| "mean_token_accuracy": 0.689780355989933, |
| "num_tokens": 18185570.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 1.3265162590891122, |
| "epoch": 1.504, |
| "grad_norm": 3.140625, |
| "learning_rate": 1.878e-05, |
| "loss": 1.2782, |
| "mean_token_accuracy": 0.6839083366096019, |
| "num_tokens": 18377903.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 1.323006783425808, |
| "epoch": 1.52, |
| "grad_norm": 3.296875, |
| "learning_rate": 1.898e-05, |
| "loss": 1.2749, |
| "mean_token_accuracy": 0.6838061761111021, |
| "num_tokens": 18579284.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 1.3307026766240597, |
| "epoch": 1.536, |
| "grad_norm": 3.515625, |
| "learning_rate": 1.918e-05, |
| "loss": 1.2627, |
| "mean_token_accuracy": 0.684997670724988, |
| "num_tokens": 18765324.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 1.2984559267759324, |
| "epoch": 1.552, |
| "grad_norm": 3.515625, |
| "learning_rate": 1.938e-05, |
| "loss": 1.2484, |
| "mean_token_accuracy": 0.6866503942757844, |
| "num_tokens": 18966382.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 1.2994249012321233, |
| "epoch": 1.568, |
| "grad_norm": 3.4375, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 1.253, |
| "mean_token_accuracy": 0.6894579119980335, |
| "num_tokens": 19160509.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 1.2866214491426944, |
| "epoch": 1.584, |
| "grad_norm": 3.171875, |
| "learning_rate": 1.978e-05, |
| "loss": 1.2375, |
| "mean_token_accuracy": 0.6898514706641435, |
| "num_tokens": 19354662.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 1.3383112341165542, |
| "epoch": 1.6, |
| "grad_norm": 3.28125, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 1.2822, |
| "mean_token_accuracy": 0.6823111370205879, |
| "num_tokens": 19543088.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_biology_entropy": 1.2766407289505004, |
| "eval_biology_loss": 1.278047800064087, |
| "eval_biology_mean_token_accuracy": 0.6927900876998901, |
| "eval_biology_num_tokens": 19543088.0, |
| "eval_biology_runtime": 39.2, |
| "eval_biology_samples_per_second": 12.755, |
| "eval_biology_steps_per_second": 3.189, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_chemistry_entropy": 0.726632200717926, |
| "eval_chemistry_loss": 0.7049177289009094, |
| "eval_chemistry_mean_token_accuracy": 0.8069638891220092, |
| "eval_chemistry_num_tokens": 19543088.0, |
| "eval_chemistry_runtime": 48.5266, |
| "eval_chemistry_samples_per_second": 10.304, |
| "eval_chemistry_steps_per_second": 2.576, |
| "step": 1000 |
| }, |
| { |
| "entropy": 1.2842272642999888, |
| "epoch": 1.616, |
| "grad_norm": 3.078125, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 1.2365, |
| "mean_token_accuracy": 0.691414737701416, |
| "num_tokens": 19743506.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 1.3330749351531268, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 3.59375, |
| "learning_rate": 1.995777777777778e-05, |
| "loss": 1.2744, |
| "mean_token_accuracy": 0.6847961049526929, |
| "num_tokens": 19937567.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 1.3074569128453732, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 2.796875, |
| "learning_rate": 1.9935555555555557e-05, |
| "loss": 1.248, |
| "mean_token_accuracy": 0.6873298797756433, |
| "num_tokens": 20129743.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 1.2786120809614658, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 3.34375, |
| "learning_rate": 1.9913333333333335e-05, |
| "loss": 1.227, |
| "mean_token_accuracy": 0.6933693554252386, |
| "num_tokens": 20331548.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 1.2929955948144198, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 3.359375, |
| "learning_rate": 1.9891111111111112e-05, |
| "loss": 1.2438, |
| "mean_token_accuracy": 0.6894369896501302, |
| "num_tokens": 20524247.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 1.2887881770730019, |
| "epoch": 1.696, |
| "grad_norm": 3.0, |
| "learning_rate": 1.986888888888889e-05, |
| "loss": 1.2341, |
| "mean_token_accuracy": 0.6910015381872654, |
| "num_tokens": 20718121.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 1.3080885756760836, |
| "epoch": 1.712, |
| "grad_norm": 3.328125, |
| "learning_rate": 1.9846666666666668e-05, |
| "loss": 1.2473, |
| "mean_token_accuracy": 0.6897755783051253, |
| "num_tokens": 20921419.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 1.2847151305526494, |
| "epoch": 1.728, |
| "grad_norm": 3.34375, |
| "learning_rate": 1.9824444444444445e-05, |
| "loss": 1.2274, |
| "mean_token_accuracy": 0.6899862293154001, |
| "num_tokens": 21118581.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 1.2512852154672145, |
| "epoch": 1.744, |
| "grad_norm": 2.90625, |
| "learning_rate": 1.9802222222222226e-05, |
| "loss": 1.2008, |
| "mean_token_accuracy": 0.6946600012481212, |
| "num_tokens": 21315128.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 1.2720022924244403, |
| "epoch": 1.76, |
| "grad_norm": 3.171875, |
| "learning_rate": 1.978e-05, |
| "loss": 1.2257, |
| "mean_token_accuracy": 0.694522064179182, |
| "num_tokens": 21511066.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_biology_entropy": 1.2505228676795959, |
| "eval_biology_loss": 1.2501847743988037, |
| "eval_biology_mean_token_accuracy": 0.6972134056091308, |
| "eval_biology_num_tokens": 21511066.0, |
| "eval_biology_runtime": 39.1908, |
| "eval_biology_samples_per_second": 12.758, |
| "eval_biology_steps_per_second": 3.19, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_chemistry_entropy": 0.7102864451408386, |
| "eval_chemistry_loss": 0.7046002149581909, |
| "eval_chemistry_mean_token_accuracy": 0.8067933526039124, |
| "eval_chemistry_num_tokens": 21511066.0, |
| "eval_chemistry_runtime": 48.815, |
| "eval_chemistry_samples_per_second": 10.243, |
| "eval_chemistry_steps_per_second": 2.561, |
| "step": 1100 |
| }, |
| { |
| "entropy": 1.281822531670332, |
| "epoch": 1.776, |
| "grad_norm": 3.375, |
| "learning_rate": 1.975777777777778e-05, |
| "loss": 1.227, |
| "mean_token_accuracy": 0.6919105790555478, |
| "num_tokens": 21704368.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 1.2503610096871853, |
| "epoch": 1.792, |
| "grad_norm": 3.03125, |
| "learning_rate": 1.9735555555555556e-05, |
| "loss": 1.1958, |
| "mean_token_accuracy": 0.6962957978248596, |
| "num_tokens": 21905784.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 1.2656195681542157, |
| "epoch": 1.808, |
| "grad_norm": 3.1875, |
| "learning_rate": 1.9713333333333337e-05, |
| "loss": 1.2172, |
| "mean_token_accuracy": 0.6964640185236931, |
| "num_tokens": 22094159.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 1.288933950662613, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 2.96875, |
| "learning_rate": 1.969111111111111e-05, |
| "loss": 1.2226, |
| "mean_token_accuracy": 0.6889266464859247, |
| "num_tokens": 22290654.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 1.3025077167898416, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 2.984375, |
| "learning_rate": 1.9668888888888892e-05, |
| "loss": 1.2518, |
| "mean_token_accuracy": 0.6867761600762605, |
| "num_tokens": 22485675.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 1.2597012262791396, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 3.125, |
| "learning_rate": 1.9646666666666666e-05, |
| "loss": 1.2159, |
| "mean_token_accuracy": 0.6929025657474994, |
| "num_tokens": 22689900.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 1.2436271466314792, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 2.65625, |
| "learning_rate": 1.9624444444444447e-05, |
| "loss": 1.1869, |
| "mean_token_accuracy": 0.700148145109415, |
| "num_tokens": 22887275.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 1.2920378752052784, |
| "epoch": 1.888, |
| "grad_norm": 3.265625, |
| "learning_rate": 1.9602222222222225e-05, |
| "loss": 1.2411, |
| "mean_token_accuracy": 0.6888992667198182, |
| "num_tokens": 23081241.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 1.2647251799702643, |
| "epoch": 1.904, |
| "grad_norm": 3.0625, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 1.2181, |
| "mean_token_accuracy": 0.693353895843029, |
| "num_tokens": 23276417.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 1.2675465818494558, |
| "epoch": 1.92, |
| "grad_norm": 3.1875, |
| "learning_rate": 1.955777777777778e-05, |
| "loss": 1.218, |
| "mean_token_accuracy": 0.6914148110896349, |
| "num_tokens": 23476340.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_biology_entropy": 1.2177199654579163, |
| "eval_biology_loss": 1.2274307012557983, |
| "eval_biology_mean_token_accuracy": 0.7003530249595642, |
| "eval_biology_num_tokens": 23476340.0, |
| "eval_biology_runtime": 39.1872, |
| "eval_biology_samples_per_second": 12.759, |
| "eval_biology_steps_per_second": 3.19, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_chemistry_entropy": 0.7087596535682679, |
| "eval_chemistry_loss": 0.7039869427680969, |
| "eval_chemistry_mean_token_accuracy": 0.8071865315437317, |
| "eval_chemistry_num_tokens": 23476340.0, |
| "eval_chemistry_runtime": 48.8046, |
| "eval_chemistry_samples_per_second": 10.245, |
| "eval_chemistry_steps_per_second": 2.561, |
| "step": 1200 |
| }, |
| { |
| "entropy": 1.2344753056764604, |
| "epoch": 1.936, |
| "grad_norm": 3.09375, |
| "learning_rate": 1.9535555555555557e-05, |
| "loss": 1.1806, |
| "mean_token_accuracy": 0.6991059686988592, |
| "num_tokens": 23673208.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 1.2617500979453324, |
| "epoch": 1.952, |
| "grad_norm": 2.953125, |
| "learning_rate": 1.9513333333333335e-05, |
| "loss": 1.2016, |
| "mean_token_accuracy": 0.6956066381186247, |
| "num_tokens": 23870993.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 1.2796590264886618, |
| "epoch": 1.968, |
| "grad_norm": 2.96875, |
| "learning_rate": 1.9491111111111113e-05, |
| "loss": 1.2352, |
| "mean_token_accuracy": 0.6916444700211286, |
| "num_tokens": 24066234.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 1.2576303217560052, |
| "epoch": 1.984, |
| "grad_norm": 3.0, |
| "learning_rate": 1.946888888888889e-05, |
| "loss": 1.2102, |
| "mean_token_accuracy": 0.6947529528290033, |
| "num_tokens": 24262389.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 1.2539904467761516, |
| "epoch": 2.0, |
| "grad_norm": 3.078125, |
| "learning_rate": 1.9446666666666668e-05, |
| "loss": 1.2133, |
| "mean_token_accuracy": 0.6965764258056879, |
| "num_tokens": 24455544.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 1.0643845289945602, |
| "epoch": 2.016, |
| "grad_norm": 3.953125, |
| "learning_rate": 1.9424444444444446e-05, |
| "loss": 1.0024, |
| "mean_token_accuracy": 0.7283362425863743, |
| "num_tokens": 24645425.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 1.0463637366890908, |
| "epoch": 2.032, |
| "grad_norm": 3.421875, |
| "learning_rate": 1.9402222222222223e-05, |
| "loss": 1.0027, |
| "mean_token_accuracy": 0.7285306710749865, |
| "num_tokens": 24835956.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 1.028846498951316, |
| "epoch": 2.048, |
| "grad_norm": 3.484375, |
| "learning_rate": 1.938e-05, |
| "loss": 0.9666, |
| "mean_token_accuracy": 0.732728997990489, |
| "num_tokens": 25036470.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 1.0426747165620327, |
| "epoch": 2.064, |
| "grad_norm": 3.3125, |
| "learning_rate": 1.935777777777778e-05, |
| "loss": 0.9818, |
| "mean_token_accuracy": 0.7311011333018541, |
| "num_tokens": 25230959.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 1.028098127245903, |
| "epoch": 2.08, |
| "grad_norm": 3.140625, |
| "learning_rate": 1.9335555555555556e-05, |
| "loss": 0.9769, |
| "mean_token_accuracy": 0.7325459264218808, |
| "num_tokens": 25422625.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_biology_entropy": 1.0245879197120666, |
| "eval_biology_loss": 1.2628997564315796, |
| "eval_biology_mean_token_accuracy": 0.6987690315246582, |
| "eval_biology_num_tokens": 25422625.0, |
| "eval_biology_runtime": 39.1891, |
| "eval_biology_samples_per_second": 12.759, |
| "eval_biology_steps_per_second": 3.19, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_chemistry_entropy": 0.5973518455028534, |
| "eval_chemistry_loss": 0.7306837439537048, |
| "eval_chemistry_mean_token_accuracy": 0.8039612555503846, |
| "eval_chemistry_num_tokens": 25422625.0, |
| "eval_chemistry_runtime": 48.8067, |
| "eval_chemistry_samples_per_second": 10.244, |
| "eval_chemistry_steps_per_second": 2.561, |
| "step": 1300 |
| }, |
| { |
| "entropy": 1.0385996278375387, |
| "epoch": 2.096, |
| "grad_norm": 3.40625, |
| "learning_rate": 1.9313333333333334e-05, |
| "loss": 0.9866, |
| "mean_token_accuracy": 0.7318900249898433, |
| "num_tokens": 25614286.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 1.0053702626377343, |
| "epoch": 2.112, |
| "grad_norm": 2.96875, |
| "learning_rate": 1.9291111111111115e-05, |
| "loss": 0.9581, |
| "mean_token_accuracy": 0.7355095498263836, |
| "num_tokens": 25810384.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 1.018695557117462, |
| "epoch": 2.128, |
| "grad_norm": 3.34375, |
| "learning_rate": 1.926888888888889e-05, |
| "loss": 0.9669, |
| "mean_token_accuracy": 0.735399067774415, |
| "num_tokens": 26005047.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 1.0224322900176048, |
| "epoch": 2.144, |
| "grad_norm": 3.734375, |
| "learning_rate": 1.924666666666667e-05, |
| "loss": 0.9788, |
| "mean_token_accuracy": 0.7329027950763702, |
| "num_tokens": 26207169.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 1.0366303831338883, |
| "epoch": 2.16, |
| "grad_norm": 3.390625, |
| "learning_rate": 1.9224444444444444e-05, |
| "loss": 0.9863, |
| "mean_token_accuracy": 0.7301317453384399, |
| "num_tokens": 26402150.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 1.0458955060690642, |
| "epoch": 2.176, |
| "grad_norm": 3.375, |
| "learning_rate": 1.9202222222222225e-05, |
| "loss": 0.999, |
| "mean_token_accuracy": 0.7267820101231337, |
| "num_tokens": 26599464.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 1.0146347057074308, |
| "epoch": 2.192, |
| "grad_norm": 3.0, |
| "learning_rate": 1.918e-05, |
| "loss": 0.9545, |
| "mean_token_accuracy": 0.7366683144122362, |
| "num_tokens": 26790601.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 1.0320534974336624, |
| "epoch": 2.208, |
| "grad_norm": 3.140625, |
| "learning_rate": 1.915777777777778e-05, |
| "loss": 0.9896, |
| "mean_token_accuracy": 0.7306183617562055, |
| "num_tokens": 26988598.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 1.061497811973095, |
| "epoch": 2.224, |
| "grad_norm": 3.171875, |
| "learning_rate": 1.9135555555555555e-05, |
| "loss": 1.0068, |
| "mean_token_accuracy": 0.7265766266733408, |
| "num_tokens": 27181933.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 1.0051178384572268, |
| "epoch": 2.24, |
| "grad_norm": 3.1875, |
| "learning_rate": 1.9113333333333336e-05, |
| "loss": 0.9516, |
| "mean_token_accuracy": 0.7369365498423577, |
| "num_tokens": 27378686.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_biology_entropy": 1.017266115665436, |
| "eval_biology_loss": 1.2557278871536255, |
| "eval_biology_mean_token_accuracy": 0.700424069404602, |
| "eval_biology_num_tokens": 27378686.0, |
| "eval_biology_runtime": 39.0925, |
| "eval_biology_samples_per_second": 12.79, |
| "eval_biology_steps_per_second": 3.198, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_chemistry_entropy": 0.5852259058952332, |
| "eval_chemistry_loss": 0.7377539277076721, |
| "eval_chemistry_mean_token_accuracy": 0.8031037306785583, |
| "eval_chemistry_num_tokens": 27378686.0, |
| "eval_chemistry_runtime": 48.69, |
| "eval_chemistry_samples_per_second": 10.269, |
| "eval_chemistry_steps_per_second": 2.567, |
| "step": 1400 |
| }, |
| { |
| "entropy": 1.0071753267198802, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 3.1875, |
| "learning_rate": 1.9091111111111113e-05, |
| "loss": 0.9674, |
| "mean_token_accuracy": 0.7368769370019436, |
| "num_tokens": 27571908.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 1.0086324002593756, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 3.203125, |
| "learning_rate": 1.906888888888889e-05, |
| "loss": 0.9656, |
| "mean_token_accuracy": 0.7324758760631085, |
| "num_tokens": 27770478.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 1.0463698647916317, |
| "epoch": 2.288, |
| "grad_norm": 3.421875, |
| "learning_rate": 1.904666666666667e-05, |
| "loss": 0.9934, |
| "mean_token_accuracy": 0.7270413629710675, |
| "num_tokens": 27965558.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 1.042240034788847, |
| "epoch": 2.304, |
| "grad_norm": 2.953125, |
| "learning_rate": 1.9024444444444446e-05, |
| "loss": 0.9889, |
| "mean_token_accuracy": 0.7297652084380388, |
| "num_tokens": 28165624.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 1.0128632839769125, |
| "epoch": 2.32, |
| "grad_norm": 3.03125, |
| "learning_rate": 1.9002222222222224e-05, |
| "loss": 0.9682, |
| "mean_token_accuracy": 0.7353047866374254, |
| "num_tokens": 28363594.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 1.021491314470768, |
| "epoch": 2.336, |
| "grad_norm": 3.375, |
| "learning_rate": 1.898e-05, |
| "loss": 0.9763, |
| "mean_token_accuracy": 0.7318463895469904, |
| "num_tokens": 28558104.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 1.0460473388433456, |
| "epoch": 2.352, |
| "grad_norm": 3.234375, |
| "learning_rate": 1.895777777777778e-05, |
| "loss": 0.9934, |
| "mean_token_accuracy": 0.7273747593164444, |
| "num_tokens": 28747265.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 1.0281270746141673, |
| "epoch": 2.368, |
| "grad_norm": 2.953125, |
| "learning_rate": 1.8935555555555556e-05, |
| "loss": 0.9686, |
| "mean_token_accuracy": 0.7312608975917101, |
| "num_tokens": 28944694.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 1.052429899200797, |
| "epoch": 2.384, |
| "grad_norm": 3.40625, |
| "learning_rate": 1.8913333333333334e-05, |
| "loss": 0.9965, |
| "mean_token_accuracy": 0.7276010267436505, |
| "num_tokens": 29136617.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 1.0189764112234116, |
| "epoch": 2.4, |
| "grad_norm": 3.25, |
| "learning_rate": 1.8891111111111115e-05, |
| "loss": 0.969, |
| "mean_token_accuracy": 0.7323620662093162, |
| "num_tokens": 29333526.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_biology_entropy": 1.0150307307243347, |
| "eval_biology_loss": 1.2410078048706055, |
| "eval_biology_mean_token_accuracy": 0.7019601583480835, |
| "eval_biology_num_tokens": 29333526.0, |
| "eval_biology_runtime": 39.111, |
| "eval_biology_samples_per_second": 12.784, |
| "eval_biology_steps_per_second": 3.196, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_chemistry_entropy": 0.5947320945262909, |
| "eval_chemistry_loss": 0.735136091709137, |
| "eval_chemistry_mean_token_accuracy": 0.8036991062164307, |
| "eval_chemistry_num_tokens": 29333526.0, |
| "eval_chemistry_runtime": 48.5118, |
| "eval_chemistry_samples_per_second": 10.307, |
| "eval_chemistry_steps_per_second": 2.577, |
| "step": 1500 |
| }, |
| { |
| "entropy": 1.0211387369781733, |
| "epoch": 2.416, |
| "grad_norm": 3.6875, |
| "learning_rate": 1.886888888888889e-05, |
| "loss": 0.9757, |
| "mean_token_accuracy": 0.7318841896951198, |
| "num_tokens": 29534153.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 1.0182897172868253, |
| "epoch": 2.432, |
| "grad_norm": 3.21875, |
| "learning_rate": 1.884666666666667e-05, |
| "loss": 0.9676, |
| "mean_token_accuracy": 0.7337216585874557, |
| "num_tokens": 29736571.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 1.055547694116831, |
| "epoch": 2.448, |
| "grad_norm": 3.65625, |
| "learning_rate": 1.8824444444444445e-05, |
| "loss": 1.0122, |
| "mean_token_accuracy": 0.7263809140771628, |
| "num_tokens": 29932423.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 1.002600496634841, |
| "epoch": 2.464, |
| "grad_norm": 3.453125, |
| "learning_rate": 1.8802222222222226e-05, |
| "loss": 0.9505, |
| "mean_token_accuracy": 0.736130004748702, |
| "num_tokens": 30137125.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 1.015967608243227, |
| "epoch": 2.48, |
| "grad_norm": 3.390625, |
| "learning_rate": 1.878e-05, |
| "loss": 0.9704, |
| "mean_token_accuracy": 0.7336820974946022, |
| "num_tokens": 30341656.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 1.0205113902688026, |
| "epoch": 2.496, |
| "grad_norm": 3.046875, |
| "learning_rate": 1.875777777777778e-05, |
| "loss": 0.9609, |
| "mean_token_accuracy": 0.7317427225410938, |
| "num_tokens": 30533732.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 1.039156510680914, |
| "epoch": 2.512, |
| "grad_norm": 3.703125, |
| "learning_rate": 1.873555555555556e-05, |
| "loss": 0.9891, |
| "mean_token_accuracy": 0.7297920010983944, |
| "num_tokens": 30730599.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 1.0211701802909374, |
| "epoch": 2.528, |
| "grad_norm": 3.34375, |
| "learning_rate": 1.8713333333333336e-05, |
| "loss": 0.9703, |
| "mean_token_accuracy": 0.7342745490372181, |
| "num_tokens": 30927950.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 1.0247683700174093, |
| "epoch": 2.544, |
| "grad_norm": 3.21875, |
| "learning_rate": 1.8691111111111114e-05, |
| "loss": 0.9847, |
| "mean_token_accuracy": 0.7314475499093532, |
| "num_tokens": 31125667.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 1.0344158336520195, |
| "epoch": 2.56, |
| "grad_norm": 3.65625, |
| "learning_rate": 1.866888888888889e-05, |
| "loss": 0.9852, |
| "mean_token_accuracy": 0.7309044182300568, |
| "num_tokens": 31315742.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_biology_entropy": 0.9934879779815674, |
| "eval_biology_loss": 1.2355761528015137, |
| "eval_biology_mean_token_accuracy": 0.703362479686737, |
| "eval_biology_num_tokens": 31315742.0, |
| "eval_biology_runtime": 39.1386, |
| "eval_biology_samples_per_second": 12.775, |
| "eval_biology_steps_per_second": 3.194, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_chemistry_entropy": 0.5818395533561707, |
| "eval_chemistry_loss": 0.7372699975967407, |
| "eval_chemistry_mean_token_accuracy": 0.8029632153511047, |
| "eval_chemistry_num_tokens": 31315742.0, |
| "eval_chemistry_runtime": 48.7641, |
| "eval_chemistry_samples_per_second": 10.253, |
| "eval_chemistry_steps_per_second": 2.563, |
| "step": 1600 |
| }, |
| { |
| "entropy": 1.0386214345693587, |
| "epoch": 2.576, |
| "grad_norm": 3.25, |
| "learning_rate": 1.864666666666667e-05, |
| "loss": 0.9889, |
| "mean_token_accuracy": 0.7297458287328482, |
| "num_tokens": 31505374.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 1.0596749667078256, |
| "epoch": 2.592, |
| "grad_norm": 3.21875, |
| "learning_rate": 1.8624444444444446e-05, |
| "loss": 1.0091, |
| "mean_token_accuracy": 0.7234568212181329, |
| "num_tokens": 31701680.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 1.043595340102911, |
| "epoch": 2.608, |
| "grad_norm": 3.140625, |
| "learning_rate": 1.8602222222222224e-05, |
| "loss": 0.9956, |
| "mean_token_accuracy": 0.7284572850912809, |
| "num_tokens": 31893221.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 1.0105513636022807, |
| "epoch": 2.624, |
| "grad_norm": 3.40625, |
| "learning_rate": 1.858e-05, |
| "loss": 0.9594, |
| "mean_token_accuracy": 0.7338220972567797, |
| "num_tokens": 32087176.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 1.0368166904896499, |
| "epoch": 2.64, |
| "grad_norm": 3.5625, |
| "learning_rate": 1.855777777777778e-05, |
| "loss": 0.9881, |
| "mean_token_accuracy": 0.7307275418192148, |
| "num_tokens": 32280689.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 1.0114492613822221, |
| "epoch": 2.656, |
| "grad_norm": 3.140625, |
| "learning_rate": 1.8535555555555557e-05, |
| "loss": 0.9589, |
| "mean_token_accuracy": 0.7375065598636865, |
| "num_tokens": 32483392.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 1.0144252099096775, |
| "epoch": 2.672, |
| "grad_norm": 3.15625, |
| "learning_rate": 1.8513333333333335e-05, |
| "loss": 0.9711, |
| "mean_token_accuracy": 0.734334035590291, |
| "num_tokens": 32683246.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 1.0010439176112413, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 3.375, |
| "learning_rate": 1.8491111111111112e-05, |
| "loss": 0.9492, |
| "mean_token_accuracy": 0.7375241097062826, |
| "num_tokens": 32884993.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 1.0266988582909107, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 3.078125, |
| "learning_rate": 1.846888888888889e-05, |
| "loss": 0.9742, |
| "mean_token_accuracy": 0.7322238199412823, |
| "num_tokens": 33088388.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.9994403839111328, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 3.453125, |
| "learning_rate": 1.8446666666666667e-05, |
| "loss": 0.9598, |
| "mean_token_accuracy": 0.7361585844308138, |
| "num_tokens": 33287208.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_biology_entropy": 1.0237988924980164, |
| "eval_biology_loss": 1.2203679084777832, |
| "eval_biology_mean_token_accuracy": 0.7062526535987854, |
| "eval_biology_num_tokens": 33287208.0, |
| "eval_biology_runtime": 39.1109, |
| "eval_biology_samples_per_second": 12.784, |
| "eval_biology_steps_per_second": 3.196, |
| "step": 1700 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_chemistry_entropy": 0.6041144397258759, |
| "eval_chemistry_loss": 0.7344766855239868, |
| "eval_chemistry_mean_token_accuracy": 0.8033582849502563, |
| "eval_chemistry_num_tokens": 33287208.0, |
| "eval_chemistry_runtime": 48.7203, |
| "eval_chemistry_samples_per_second": 10.263, |
| "eval_chemistry_steps_per_second": 2.566, |
| "step": 1700 |
| }, |
| { |
| "entropy": 1.0407861892133952, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 3.09375, |
| "learning_rate": 1.842444444444445e-05, |
| "loss": 0.9879, |
| "mean_token_accuracy": 0.7298587996512651, |
| "num_tokens": 33474750.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 1.0247716046869755, |
| "epoch": 2.752, |
| "grad_norm": 3.609375, |
| "learning_rate": 1.8402222222222223e-05, |
| "loss": 0.9769, |
| "mean_token_accuracy": 0.730589097738266, |
| "num_tokens": 33664302.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 1.0562083158642053, |
| "epoch": 2.768, |
| "grad_norm": 3.390625, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 1.0054, |
| "mean_token_accuracy": 0.7239043671637774, |
| "num_tokens": 33854081.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 1.016645783931017, |
| "epoch": 2.784, |
| "grad_norm": 3.0, |
| "learning_rate": 1.8357777777777778e-05, |
| "loss": 0.9691, |
| "mean_token_accuracy": 0.7346661690622568, |
| "num_tokens": 34053149.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 1.0153463836759329, |
| "epoch": 2.8, |
| "grad_norm": 3.125, |
| "learning_rate": 1.833555555555556e-05, |
| "loss": 0.969, |
| "mean_token_accuracy": 0.7363858807832002, |
| "num_tokens": 34254105.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 1.0121935941278934, |
| "epoch": 2.816, |
| "grad_norm": 3.15625, |
| "learning_rate": 1.8313333333333333e-05, |
| "loss": 0.9579, |
| "mean_token_accuracy": 0.7347628369927406, |
| "num_tokens": 34453342.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 1.050192491337657, |
| "epoch": 2.832, |
| "grad_norm": 3.421875, |
| "learning_rate": 1.8291111111111114e-05, |
| "loss": 0.9938, |
| "mean_token_accuracy": 0.7286655321717262, |
| "num_tokens": 34644113.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 1.0338923085480929, |
| "epoch": 2.848, |
| "grad_norm": 3.296875, |
| "learning_rate": 1.8268888888888888e-05, |
| "loss": 0.9888, |
| "mean_token_accuracy": 0.7316403832286597, |
| "num_tokens": 34835062.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 1.0176855057477951, |
| "epoch": 2.864, |
| "grad_norm": 3.25, |
| "learning_rate": 1.824666666666667e-05, |
| "loss": 0.9754, |
| "mean_token_accuracy": 0.7330775160342455, |
| "num_tokens": 35032361.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 1.0399887025356294, |
| "epoch": 2.88, |
| "grad_norm": 3.21875, |
| "learning_rate": 1.8224444444444447e-05, |
| "loss": 0.9921, |
| "mean_token_accuracy": 0.7296446476131677, |
| "num_tokens": 35226653.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_biology_entropy": 1.0291271986961366, |
| "eval_biology_loss": 1.2109084129333496, |
| "eval_biology_mean_token_accuracy": 0.70716144323349, |
| "eval_biology_num_tokens": 35226653.0, |
| "eval_biology_runtime": 39.2084, |
| "eval_biology_samples_per_second": 12.752, |
| "eval_biology_steps_per_second": 3.188, |
| "step": 1800 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_chemistry_entropy": 0.5928264305591583, |
| "eval_chemistry_loss": 0.7367877960205078, |
| "eval_chemistry_mean_token_accuracy": 0.8034037079811096, |
| "eval_chemistry_num_tokens": 35226653.0, |
| "eval_chemistry_runtime": 48.8018, |
| "eval_chemistry_samples_per_second": 10.246, |
| "eval_chemistry_steps_per_second": 2.561, |
| "step": 1800 |
| }, |
| { |
| "entropy": 1.0336259830743075, |
| "epoch": 2.896, |
| "grad_norm": 3.234375, |
| "learning_rate": 1.8202222222222225e-05, |
| "loss": 0.9812, |
| "mean_token_accuracy": 0.7294993791729212, |
| "num_tokens": 35421873.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 1.0508763272315265, |
| "epoch": 2.912, |
| "grad_norm": 3.5625, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 1.0003, |
| "mean_token_accuracy": 0.728679533302784, |
| "num_tokens": 35614590.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 1.0313632506877184, |
| "epoch": 2.928, |
| "grad_norm": 3.515625, |
| "learning_rate": 1.815777777777778e-05, |
| "loss": 0.9877, |
| "mean_token_accuracy": 0.7299741402268409, |
| "num_tokens": 35806494.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 1.023673979192972, |
| "epoch": 2.944, |
| "grad_norm": 3.09375, |
| "learning_rate": 1.8135555555555557e-05, |
| "loss": 0.9814, |
| "mean_token_accuracy": 0.7329835917800664, |
| "num_tokens": 36001867.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 1.0165797971189021, |
| "epoch": 2.96, |
| "grad_norm": 3.296875, |
| "learning_rate": 1.8113333333333335e-05, |
| "loss": 0.9588, |
| "mean_token_accuracy": 0.7343671884387731, |
| "num_tokens": 36198806.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 1.0112668476998805, |
| "epoch": 2.976, |
| "grad_norm": 3.1875, |
| "learning_rate": 1.8091111111111113e-05, |
| "loss": 0.9622, |
| "mean_token_accuracy": 0.7349605076014996, |
| "num_tokens": 36393098.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 1.0263954121619463, |
| "epoch": 2.992, |
| "grad_norm": 3.265625, |
| "learning_rate": 1.806888888888889e-05, |
| "loss": 0.9775, |
| "mean_token_accuracy": 0.7333775267004967, |
| "num_tokens": 36589074.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.938998518884182, |
| "epoch": 3.008, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.8046666666666668e-05, |
| "loss": 0.8348, |
| "mean_token_accuracy": 0.7635289087891579, |
| "num_tokens": 36778710.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.673611904680729, |
| "epoch": 3.024, |
| "grad_norm": 3.609375, |
| "learning_rate": 1.8024444444444445e-05, |
| "loss": 0.6431, |
| "mean_token_accuracy": 0.8048487104475498, |
| "num_tokens": 36966789.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.6563162552192807, |
| "epoch": 3.04, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.8002222222222223e-05, |
| "loss": 0.598, |
| "mean_token_accuracy": 0.8154135558754205, |
| "num_tokens": 37156529.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_biology_entropy": 0.7205301468372345, |
| "eval_biology_loss": 1.454393744468689, |
| "eval_biology_mean_token_accuracy": 0.6943806533813477, |
| "eval_biology_num_tokens": 37156529.0, |
| "eval_biology_runtime": 39.1135, |
| "eval_biology_samples_per_second": 12.783, |
| "eval_biology_steps_per_second": 3.196, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_chemistry_entropy": 0.44404922652244566, |
| "eval_chemistry_loss": 0.8767896890640259, |
| "eval_chemistry_mean_token_accuracy": 0.7932383036613464, |
| "eval_chemistry_num_tokens": 37156529.0, |
| "eval_chemistry_runtime": 48.7267, |
| "eval_chemistry_samples_per_second": 10.261, |
| "eval_chemistry_steps_per_second": 2.565, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.6639188259840012, |
| "epoch": 3.056, |
| "grad_norm": 4.125, |
| "learning_rate": 1.798e-05, |
| "loss": 0.6032, |
| "mean_token_accuracy": 0.8127994637936353, |
| "num_tokens": 37351325.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.6422042794525623, |
| "epoch": 3.072, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.7957777777777778e-05, |
| "loss": 0.5921, |
| "mean_token_accuracy": 0.8172261964529752, |
| "num_tokens": 37541069.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.6540521441027523, |
| "epoch": 3.088, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.7935555555555556e-05, |
| "loss": 0.5993, |
| "mean_token_accuracy": 0.8149240974336862, |
| "num_tokens": 37733440.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.6463017119094729, |
| "epoch": 3.104, |
| "grad_norm": 3.71875, |
| "learning_rate": 1.7913333333333337e-05, |
| "loss": 0.5868, |
| "mean_token_accuracy": 0.8175160981714725, |
| "num_tokens": 37933661.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.6483396209776402, |
| "epoch": 3.12, |
| "grad_norm": 4.0, |
| "learning_rate": 1.789111111111111e-05, |
| "loss": 0.6017, |
| "mean_token_accuracy": 0.8143340889364481, |
| "num_tokens": 38126905.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.6592650156468153, |
| "epoch": 3.136, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.7868888888888892e-05, |
| "loss": 0.598, |
| "mean_token_accuracy": 0.8149209085851907, |
| "num_tokens": 38316042.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.6408368192613125, |
| "epoch": 3.152, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.7846666666666666e-05, |
| "loss": 0.5898, |
| "mean_token_accuracy": 0.8172995913773775, |
| "num_tokens": 38514212.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.6699947357177735, |
| "epoch": 3.168, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.7824444444444447e-05, |
| "loss": 0.6136, |
| "mean_token_accuracy": 0.8107931781560183, |
| "num_tokens": 38708822.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.6612202525138855, |
| "epoch": 3.184, |
| "grad_norm": 3.859375, |
| "learning_rate": 1.780222222222222e-05, |
| "loss": 0.6107, |
| "mean_token_accuracy": 0.8108679629862309, |
| "num_tokens": 38900708.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.6577222358435393, |
| "epoch": 3.2, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 0.6033, |
| "mean_token_accuracy": 0.8127841059118509, |
| "num_tokens": 39102325.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_biology_entropy": 0.7194305346012115, |
| "eval_biology_loss": 1.458989143371582, |
| "eval_biology_mean_token_accuracy": 0.6955648012161255, |
| "eval_biology_num_tokens": 39102325.0, |
| "eval_biology_runtime": 39.0915, |
| "eval_biology_samples_per_second": 12.791, |
| "eval_biology_steps_per_second": 3.198, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_chemistry_entropy": 0.4279054062366486, |
| "eval_chemistry_loss": 0.8939092755317688, |
| "eval_chemistry_mean_token_accuracy": 0.7933331608772278, |
| "eval_chemistry_num_tokens": 39102325.0, |
| "eval_chemistry_runtime": 48.5319, |
| "eval_chemistry_samples_per_second": 10.303, |
| "eval_chemistry_steps_per_second": 2.576, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 10000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 16, |
| "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": 3.279038413131436e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
