| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 8.0, |
| "eval_steps": 100, |
| "global_step": 5000, |
| "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 |
| }, |
| { |
| "entropy": 0.6628522459417582, |
| "epoch": 3.216, |
| "grad_norm": 4.5, |
| "learning_rate": 1.7757777777777777e-05, |
| "loss": 0.6056, |
| "mean_token_accuracy": 0.8133372392505407, |
| "num_tokens": 39295913.0, |
| "step": 2010 |
| }, |
| { |
| "entropy": 0.6509159937500953, |
| "epoch": 3.232, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.7735555555555558e-05, |
| "loss": 0.5956, |
| "mean_token_accuracy": 0.8132004003971816, |
| "num_tokens": 39495778.0, |
| "step": 2020 |
| }, |
| { |
| "entropy": 0.6906503856182098, |
| "epoch": 3.248, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.7713333333333335e-05, |
| "loss": 0.6307, |
| "mean_token_accuracy": 0.8041897661983967, |
| "num_tokens": 39681539.0, |
| "step": 2030 |
| }, |
| { |
| "entropy": 0.6459178572520614, |
| "epoch": 3.2640000000000002, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.7691111111111113e-05, |
| "loss": 0.5894, |
| "mean_token_accuracy": 0.8152297649532556, |
| "num_tokens": 39887936.0, |
| "step": 2040 |
| }, |
| { |
| "entropy": 0.6582429932430387, |
| "epoch": 3.2800000000000002, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.766888888888889e-05, |
| "loss": 0.6037, |
| "mean_token_accuracy": 0.8127248145639896, |
| "num_tokens": 40081612.0, |
| "step": 2050 |
| }, |
| { |
| "entropy": 0.6647687572985888, |
| "epoch": 3.296, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.7646666666666668e-05, |
| "loss": 0.6134, |
| "mean_token_accuracy": 0.8107028409838677, |
| "num_tokens": 40274536.0, |
| "step": 2060 |
| }, |
| { |
| "entropy": 0.6784954000264406, |
| "epoch": 3.312, |
| "grad_norm": 4.5, |
| "learning_rate": 1.7624444444444446e-05, |
| "loss": 0.6277, |
| "mean_token_accuracy": 0.8066128794103861, |
| "num_tokens": 40465691.0, |
| "step": 2070 |
| }, |
| { |
| "entropy": 0.6526031637564301, |
| "epoch": 3.328, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.7602222222222223e-05, |
| "loss": 0.5965, |
| "mean_token_accuracy": 0.81493533924222, |
| "num_tokens": 40665740.0, |
| "step": 2080 |
| }, |
| { |
| "entropy": 0.6589240668341517, |
| "epoch": 3.344, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.758e-05, |
| "loss": 0.6031, |
| "mean_token_accuracy": 0.8119147256016731, |
| "num_tokens": 40859755.0, |
| "step": 2090 |
| }, |
| { |
| "entropy": 0.6566657887771725, |
| "epoch": 3.36, |
| "grad_norm": 3.984375, |
| "learning_rate": 1.755777777777778e-05, |
| "loss": 0.6065, |
| "mean_token_accuracy": 0.8120837610214948, |
| "num_tokens": 41059360.0, |
| "step": 2100 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_biology_entropy": 0.7223304212093353, |
| "eval_biology_loss": 1.4520024061203003, |
| "eval_biology_mean_token_accuracy": 0.6971806507110596, |
| "eval_biology_num_tokens": 41059360.0, |
| "eval_biology_runtime": 39.0816, |
| "eval_biology_samples_per_second": 12.794, |
| "eval_biology_steps_per_second": 3.198, |
| "step": 2100 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_chemistry_entropy": 0.425049998998642, |
| "eval_chemistry_loss": 0.8979185223579407, |
| "eval_chemistry_mean_token_accuracy": 0.7932314896583557, |
| "eval_chemistry_num_tokens": 41059360.0, |
| "eval_chemistry_runtime": 48.685, |
| "eval_chemistry_samples_per_second": 10.27, |
| "eval_chemistry_steps_per_second": 2.568, |
| "step": 2100 |
| }, |
| { |
| "entropy": 0.6639438619837165, |
| "epoch": 3.376, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.7535555555555556e-05, |
| "loss": 0.6149, |
| "mean_token_accuracy": 0.8104340396821499, |
| "num_tokens": 41256224.0, |
| "step": 2110 |
| }, |
| { |
| "entropy": 0.6947120493277907, |
| "epoch": 3.392, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.7513333333333334e-05, |
| "loss": 0.6371, |
| "mean_token_accuracy": 0.8046501569449902, |
| "num_tokens": 41449411.0, |
| "step": 2120 |
| }, |
| { |
| "entropy": 0.6828485745936632, |
| "epoch": 3.408, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.749111111111111e-05, |
| "loss": 0.6288, |
| "mean_token_accuracy": 0.8066561881452798, |
| "num_tokens": 41636184.0, |
| "step": 2130 |
| }, |
| { |
| "entropy": 0.6725861068814993, |
| "epoch": 3.424, |
| "grad_norm": 3.90625, |
| "learning_rate": 1.746888888888889e-05, |
| "loss": 0.6202, |
| "mean_token_accuracy": 0.81025187112391, |
| "num_tokens": 41831661.0, |
| "step": 2140 |
| }, |
| { |
| "entropy": 0.6702702101320028, |
| "epoch": 3.44, |
| "grad_norm": 4.5, |
| "learning_rate": 1.7446666666666667e-05, |
| "loss": 0.6107, |
| "mean_token_accuracy": 0.8081615891307592, |
| "num_tokens": 42030546.0, |
| "step": 2150 |
| }, |
| { |
| "entropy": 0.6535747822374105, |
| "epoch": 3.456, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.7424444444444444e-05, |
| "loss": 0.6064, |
| "mean_token_accuracy": 0.8117943584918976, |
| "num_tokens": 42232724.0, |
| "step": 2160 |
| }, |
| { |
| "entropy": 0.6701465247198939, |
| "epoch": 3.472, |
| "grad_norm": 4.25, |
| "learning_rate": 1.7402222222222222e-05, |
| "loss": 0.6144, |
| "mean_token_accuracy": 0.8089125752449036, |
| "num_tokens": 42428139.0, |
| "step": 2170 |
| }, |
| { |
| "entropy": 0.6602701621130109, |
| "epoch": 3.488, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 0.6051, |
| "mean_token_accuracy": 0.8109641902148723, |
| "num_tokens": 42624281.0, |
| "step": 2180 |
| }, |
| { |
| "entropy": 0.6750716870650649, |
| "epoch": 3.504, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.735777777777778e-05, |
| "loss": 0.6224, |
| "mean_token_accuracy": 0.8076240297406911, |
| "num_tokens": 42820999.0, |
| "step": 2190 |
| }, |
| { |
| "entropy": 0.6583628479391337, |
| "epoch": 3.52, |
| "grad_norm": 4.25, |
| "learning_rate": 1.7335555555555558e-05, |
| "loss": 0.6027, |
| "mean_token_accuracy": 0.8134463246911764, |
| "num_tokens": 43024350.0, |
| "step": 2200 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_biology_entropy": 0.7312813591957092, |
| "eval_biology_loss": 1.448984146118164, |
| "eval_biology_mean_token_accuracy": 0.6975588989257813, |
| "eval_biology_num_tokens": 43024350.0, |
| "eval_biology_runtime": 39.1828, |
| "eval_biology_samples_per_second": 12.761, |
| "eval_biology_steps_per_second": 3.19, |
| "step": 2200 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_chemistry_entropy": 0.42386009073257447, |
| "eval_chemistry_loss": 0.9041973352432251, |
| "eval_chemistry_mean_token_accuracy": 0.7924366879463196, |
| "eval_chemistry_num_tokens": 43024350.0, |
| "eval_chemistry_runtime": 48.8105, |
| "eval_chemistry_samples_per_second": 10.244, |
| "eval_chemistry_steps_per_second": 2.561, |
| "step": 2200 |
| }, |
| { |
| "entropy": 0.6752564821392297, |
| "epoch": 3.536, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.7313333333333336e-05, |
| "loss": 0.6237, |
| "mean_token_accuracy": 0.8093563176691532, |
| "num_tokens": 43214925.0, |
| "step": 2210 |
| }, |
| { |
| "entropy": 0.675455342233181, |
| "epoch": 3.552, |
| "grad_norm": 4.375, |
| "learning_rate": 1.7291111111111113e-05, |
| "loss": 0.6187, |
| "mean_token_accuracy": 0.8095767885446549, |
| "num_tokens": 43409227.0, |
| "step": 2220 |
| }, |
| { |
| "entropy": 0.6612193696200848, |
| "epoch": 3.568, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.726888888888889e-05, |
| "loss": 0.6128, |
| "mean_token_accuracy": 0.8100322619080543, |
| "num_tokens": 43609101.0, |
| "step": 2230 |
| }, |
| { |
| "entropy": 0.6645467173308134, |
| "epoch": 3.584, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.724666666666667e-05, |
| "loss": 0.61, |
| "mean_token_accuracy": 0.8110004521906375, |
| "num_tokens": 43806883.0, |
| "step": 2240 |
| }, |
| { |
| "entropy": 0.6700274987146259, |
| "epoch": 3.6, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.7224444444444446e-05, |
| "loss": 0.6151, |
| "mean_token_accuracy": 0.8092441029846669, |
| "num_tokens": 43996466.0, |
| "step": 2250 |
| }, |
| { |
| "entropy": 0.6563864570111037, |
| "epoch": 3.616, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.7202222222222224e-05, |
| "loss": 0.6092, |
| "mean_token_accuracy": 0.8101060092449188, |
| "num_tokens": 44199791.0, |
| "step": 2260 |
| }, |
| { |
| "entropy": 0.6729124957695604, |
| "epoch": 3.632, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.718e-05, |
| "loss": 0.6165, |
| "mean_token_accuracy": 0.8097267791628837, |
| "num_tokens": 44400793.0, |
| "step": 2270 |
| }, |
| { |
| "entropy": 0.7002201689407229, |
| "epoch": 3.648, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.715777777777778e-05, |
| "loss": 0.6449, |
| "mean_token_accuracy": 0.8013300366699696, |
| "num_tokens": 44592807.0, |
| "step": 2280 |
| }, |
| { |
| "entropy": 0.6589280311018229, |
| "epoch": 3.664, |
| "grad_norm": 4.125, |
| "learning_rate": 1.7135555555555557e-05, |
| "loss": 0.6106, |
| "mean_token_accuracy": 0.8104540213942528, |
| "num_tokens": 44796238.0, |
| "step": 2290 |
| }, |
| { |
| "entropy": 0.6718129109591245, |
| "epoch": 3.68, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.7113333333333334e-05, |
| "loss": 0.6148, |
| "mean_token_accuracy": 0.808036233112216, |
| "num_tokens": 44996201.0, |
| "step": 2300 |
| }, |
| { |
| "epoch": 3.68, |
| "eval_biology_entropy": 0.7342447080612182, |
| "eval_biology_loss": 1.4424549341201782, |
| "eval_biology_mean_token_accuracy": 0.6978958992958069, |
| "eval_biology_num_tokens": 44996201.0, |
| "eval_biology_runtime": 39.18, |
| "eval_biology_samples_per_second": 12.762, |
| "eval_biology_steps_per_second": 3.19, |
| "step": 2300 |
| }, |
| { |
| "epoch": 3.68, |
| "eval_chemistry_entropy": 0.42464341497421265, |
| "eval_chemistry_loss": 0.9024158120155334, |
| "eval_chemistry_mean_token_accuracy": 0.7931222615242004, |
| "eval_chemistry_num_tokens": 44996201.0, |
| "eval_chemistry_runtime": 48.8209, |
| "eval_chemistry_samples_per_second": 10.242, |
| "eval_chemistry_steps_per_second": 2.56, |
| "step": 2300 |
| }, |
| { |
| "entropy": 0.6750511961057782, |
| "epoch": 3.6959999999999997, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.7091111111111112e-05, |
| "loss": 0.6263, |
| "mean_token_accuracy": 0.8063635263592005, |
| "num_tokens": 45195456.0, |
| "step": 2310 |
| }, |
| { |
| "entropy": 0.6969444552436471, |
| "epoch": 3.7119999999999997, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.706888888888889e-05, |
| "loss": 0.6355, |
| "mean_token_accuracy": 0.8052191227674484, |
| "num_tokens": 45385114.0, |
| "step": 2320 |
| }, |
| { |
| "entropy": 0.6575890483334661, |
| "epoch": 3.7279999999999998, |
| "grad_norm": 3.96875, |
| "learning_rate": 1.704666666666667e-05, |
| "loss": 0.6079, |
| "mean_token_accuracy": 0.8101479094475508, |
| "num_tokens": 45586032.0, |
| "step": 2330 |
| }, |
| { |
| "entropy": 0.6878007039427757, |
| "epoch": 3.7439999999999998, |
| "grad_norm": 4.25, |
| "learning_rate": 1.7024444444444445e-05, |
| "loss": 0.6287, |
| "mean_token_accuracy": 0.8045054972171783, |
| "num_tokens": 45777138.0, |
| "step": 2340 |
| }, |
| { |
| "entropy": 0.6799204269424081, |
| "epoch": 3.76, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.7002222222222226e-05, |
| "loss": 0.6291, |
| "mean_token_accuracy": 0.8070768497884273, |
| "num_tokens": 45971818.0, |
| "step": 2350 |
| }, |
| { |
| "entropy": 0.6708019917830825, |
| "epoch": 3.776, |
| "grad_norm": 4.125, |
| "learning_rate": 1.698e-05, |
| "loss": 0.6237, |
| "mean_token_accuracy": 0.8083038441836834, |
| "num_tokens": 46167844.0, |
| "step": 2360 |
| }, |
| { |
| "entropy": 0.6993245150893926, |
| "epoch": 3.792, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.695777777777778e-05, |
| "loss": 0.6447, |
| "mean_token_accuracy": 0.8005988635122776, |
| "num_tokens": 46362538.0, |
| "step": 2370 |
| }, |
| { |
| "entropy": 0.6768058065325022, |
| "epoch": 3.808, |
| "grad_norm": 4.25, |
| "learning_rate": 1.6935555555555555e-05, |
| "loss": 0.6247, |
| "mean_token_accuracy": 0.8075179129838943, |
| "num_tokens": 46553166.0, |
| "step": 2380 |
| }, |
| { |
| "entropy": 0.6853747364133597, |
| "epoch": 3.824, |
| "grad_norm": 4.125, |
| "learning_rate": 1.6913333333333336e-05, |
| "loss": 0.635, |
| "mean_token_accuracy": 0.8052531398832798, |
| "num_tokens": 46744198.0, |
| "step": 2390 |
| }, |
| { |
| "entropy": 0.6714792435988783, |
| "epoch": 3.84, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.689111111111111e-05, |
| "loss": 0.6228, |
| "mean_token_accuracy": 0.8093580879271031, |
| "num_tokens": 46941095.0, |
| "step": 2400 |
| }, |
| { |
| "epoch": 3.84, |
| "eval_biology_entropy": 0.7420620145797729, |
| "eval_biology_loss": 1.4358100891113281, |
| "eval_biology_mean_token_accuracy": 0.698259696006775, |
| "eval_biology_num_tokens": 46941095.0, |
| "eval_biology_runtime": 39.0623, |
| "eval_biology_samples_per_second": 12.8, |
| "eval_biology_steps_per_second": 3.2, |
| "step": 2400 |
| }, |
| { |
| "epoch": 3.84, |
| "eval_chemistry_entropy": 0.4296411466598511, |
| "eval_chemistry_loss": 0.902924656867981, |
| "eval_chemistry_mean_token_accuracy": 0.7924109597206116, |
| "eval_chemistry_num_tokens": 46941095.0, |
| "eval_chemistry_runtime": 48.6871, |
| "eval_chemistry_samples_per_second": 10.27, |
| "eval_chemistry_steps_per_second": 2.567, |
| "step": 2400 |
| }, |
| { |
| "entropy": 0.6763832990080119, |
| "epoch": 3.856, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.686888888888889e-05, |
| "loss": 0.622, |
| "mean_token_accuracy": 0.8078614696860313, |
| "num_tokens": 47139983.0, |
| "step": 2410 |
| }, |
| { |
| "entropy": 0.690857807919383, |
| "epoch": 3.872, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.684666666666667e-05, |
| "loss": 0.6425, |
| "mean_token_accuracy": 0.8026266809552908, |
| "num_tokens": 47337119.0, |
| "step": 2420 |
| }, |
| { |
| "entropy": 0.6782918659970164, |
| "epoch": 3.888, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.6824444444444447e-05, |
| "loss": 0.6103, |
| "mean_token_accuracy": 0.8091849368065596, |
| "num_tokens": 47535142.0, |
| "step": 2430 |
| }, |
| { |
| "entropy": 0.6686505619436502, |
| "epoch": 3.904, |
| "grad_norm": 4.25, |
| "learning_rate": 1.6802222222222224e-05, |
| "loss": 0.6245, |
| "mean_token_accuracy": 0.8080017164349556, |
| "num_tokens": 47729505.0, |
| "step": 2440 |
| }, |
| { |
| "entropy": 0.6681339140981436, |
| "epoch": 3.92, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 0.6189, |
| "mean_token_accuracy": 0.8098367966711522, |
| "num_tokens": 47929285.0, |
| "step": 2450 |
| }, |
| { |
| "entropy": 0.6815527578815818, |
| "epoch": 3.936, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.675777777777778e-05, |
| "loss": 0.6279, |
| "mean_token_accuracy": 0.8059242676943541, |
| "num_tokens": 48127866.0, |
| "step": 2460 |
| }, |
| { |
| "entropy": 0.6731824316084385, |
| "epoch": 3.952, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.6735555555555557e-05, |
| "loss": 0.6223, |
| "mean_token_accuracy": 0.8080686226487159, |
| "num_tokens": 48325429.0, |
| "step": 2470 |
| }, |
| { |
| "entropy": 0.7055601105093956, |
| "epoch": 3.968, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.6713333333333335e-05, |
| "loss": 0.6452, |
| "mean_token_accuracy": 0.8004195280373096, |
| "num_tokens": 48522725.0, |
| "step": 2480 |
| }, |
| { |
| "entropy": 0.6725156113505364, |
| "epoch": 3.984, |
| "grad_norm": 3.921875, |
| "learning_rate": 1.6691111111111112e-05, |
| "loss": 0.6208, |
| "mean_token_accuracy": 0.8087173383682966, |
| "num_tokens": 48719975.0, |
| "step": 2490 |
| }, |
| { |
| "entropy": 0.6917624410241843, |
| "epoch": 4.0, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.666888888888889e-05, |
| "loss": 0.6448, |
| "mean_token_accuracy": 0.8019643574953079, |
| "num_tokens": 48911088.0, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_biology_entropy": 0.7488004591464996, |
| "eval_biology_loss": 1.4288437366485596, |
| "eval_biology_mean_token_accuracy": 0.6984080262184144, |
| "eval_biology_num_tokens": 48911088.0, |
| "eval_biology_runtime": 39.1871, |
| "eval_biology_samples_per_second": 12.759, |
| "eval_biology_steps_per_second": 3.19, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_chemistry_entropy": 0.43333416879177095, |
| "eval_chemistry_loss": 0.9011425971984863, |
| "eval_chemistry_mean_token_accuracy": 0.7929084548950195, |
| "eval_chemistry_num_tokens": 48911088.0, |
| "eval_chemistry_runtime": 48.521, |
| "eval_chemistry_samples_per_second": 10.305, |
| "eval_chemistry_steps_per_second": 2.576, |
| "step": 2500 |
| }, |
| { |
| "entropy": 0.4309111710637808, |
| "epoch": 4.016, |
| "grad_norm": 5.40625, |
| "learning_rate": 1.6646666666666668e-05, |
| "loss": 0.3163, |
| "mean_token_accuracy": 0.9044911481440068, |
| "num_tokens": 49109596.0, |
| "step": 2510 |
| }, |
| { |
| "entropy": 0.3868551283143461, |
| "epoch": 4.032, |
| "grad_norm": 3.609375, |
| "learning_rate": 1.6624444444444445e-05, |
| "loss": 0.3027, |
| "mean_token_accuracy": 0.906030959635973, |
| "num_tokens": 49308883.0, |
| "step": 2520 |
| }, |
| { |
| "entropy": 0.38658891851082444, |
| "epoch": 4.048, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.6602222222222223e-05, |
| "loss": 0.2897, |
| "mean_token_accuracy": 0.908681920170784, |
| "num_tokens": 49506541.0, |
| "step": 2530 |
| }, |
| { |
| "entropy": 0.3816244132816792, |
| "epoch": 4.064, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.658e-05, |
| "loss": 0.2942, |
| "mean_token_accuracy": 0.9073891270905733, |
| "num_tokens": 49696752.0, |
| "step": 2540 |
| }, |
| { |
| "entropy": 0.3789869106374681, |
| "epoch": 4.08, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.6557777777777778e-05, |
| "loss": 0.2894, |
| "mean_token_accuracy": 0.9080625347793102, |
| "num_tokens": 49893113.0, |
| "step": 2550 |
| }, |
| { |
| "entropy": 0.3826861511915922, |
| "epoch": 4.096, |
| "grad_norm": 3.5, |
| "learning_rate": 1.6535555555555556e-05, |
| "loss": 0.2966, |
| "mean_token_accuracy": 0.9070118863135577, |
| "num_tokens": 50083855.0, |
| "step": 2560 |
| }, |
| { |
| "entropy": 0.38291181391105056, |
| "epoch": 4.112, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.6513333333333333e-05, |
| "loss": 0.2926, |
| "mean_token_accuracy": 0.9068987872451544, |
| "num_tokens": 50278886.0, |
| "step": 2570 |
| }, |
| { |
| "entropy": 0.38163715079426763, |
| "epoch": 4.128, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.6491111111111114e-05, |
| "loss": 0.2923, |
| "mean_token_accuracy": 0.9077801682054997, |
| "num_tokens": 50471344.0, |
| "step": 2580 |
| }, |
| { |
| "entropy": 0.3836596327833831, |
| "epoch": 4.144, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.646888888888889e-05, |
| "loss": 0.2958, |
| "mean_token_accuracy": 0.9058610606938601, |
| "num_tokens": 50658788.0, |
| "step": 2590 |
| }, |
| { |
| "entropy": 0.3837190412916243, |
| "epoch": 4.16, |
| "grad_norm": 3.796875, |
| "learning_rate": 1.644666666666667e-05, |
| "loss": 0.2974, |
| "mean_token_accuracy": 0.905021695420146, |
| "num_tokens": 50853770.0, |
| "step": 2600 |
| }, |
| { |
| "epoch": 4.16, |
| "eval_biology_entropy": 0.5419651811122894, |
| "eval_biology_loss": 1.7417488098144531, |
| "eval_biology_mean_token_accuracy": 0.6883382601737976, |
| "eval_biology_num_tokens": 50853770.0, |
| "eval_biology_runtime": 39.1152, |
| "eval_biology_samples_per_second": 12.783, |
| "eval_biology_steps_per_second": 3.196, |
| "step": 2600 |
| }, |
| { |
| "epoch": 4.16, |
| "eval_chemistry_entropy": 0.3343317413330078, |
| "eval_chemistry_loss": 1.0794504880905151, |
| "eval_chemistry_mean_token_accuracy": 0.7853766241073609, |
| "eval_chemistry_num_tokens": 50853770.0, |
| "eval_chemistry_runtime": 48.7311, |
| "eval_chemistry_samples_per_second": 10.26, |
| "eval_chemistry_steps_per_second": 2.565, |
| "step": 2600 |
| }, |
| { |
| "entropy": 0.37414240883663297, |
| "epoch": 4.176, |
| "grad_norm": 3.84375, |
| "learning_rate": 1.6424444444444444e-05, |
| "loss": 0.2893, |
| "mean_token_accuracy": 0.9082878790795803, |
| "num_tokens": 51053592.0, |
| "step": 2610 |
| }, |
| { |
| "entropy": 0.3877230207435787, |
| "epoch": 4.192, |
| "grad_norm": 4.125, |
| "learning_rate": 1.6402222222222225e-05, |
| "loss": 0.2981, |
| "mean_token_accuracy": 0.9046992681920528, |
| "num_tokens": 51245972.0, |
| "step": 2620 |
| }, |
| { |
| "entropy": 0.37879814049229027, |
| "epoch": 4.208, |
| "grad_norm": 3.71875, |
| "learning_rate": 1.638e-05, |
| "loss": 0.2988, |
| "mean_token_accuracy": 0.9060565620660782, |
| "num_tokens": 51445425.0, |
| "step": 2630 |
| }, |
| { |
| "entropy": 0.38125489223748443, |
| "epoch": 4.224, |
| "grad_norm": 3.9375, |
| "learning_rate": 1.635777777777778e-05, |
| "loss": 0.2953, |
| "mean_token_accuracy": 0.9068669069558382, |
| "num_tokens": 51642750.0, |
| "step": 2640 |
| }, |
| { |
| "entropy": 0.37129258885979655, |
| "epoch": 4.24, |
| "grad_norm": 4.125, |
| "learning_rate": 1.6335555555555558e-05, |
| "loss": 0.2908, |
| "mean_token_accuracy": 0.9074220947921277, |
| "num_tokens": 51842760.0, |
| "step": 2650 |
| }, |
| { |
| "entropy": 0.3658179035410285, |
| "epoch": 4.256, |
| "grad_norm": 3.921875, |
| "learning_rate": 1.6313333333333335e-05, |
| "loss": 0.2849, |
| "mean_token_accuracy": 0.9101897101849318, |
| "num_tokens": 52045869.0, |
| "step": 2660 |
| }, |
| { |
| "entropy": 0.37624444244429467, |
| "epoch": 4.272, |
| "grad_norm": 3.84375, |
| "learning_rate": 1.6291111111111113e-05, |
| "loss": 0.2952, |
| "mean_token_accuracy": 0.9070649083703757, |
| "num_tokens": 52241862.0, |
| "step": 2670 |
| }, |
| { |
| "entropy": 0.3684785286895931, |
| "epoch": 4.288, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.626888888888889e-05, |
| "loss": 0.2924, |
| "mean_token_accuracy": 0.9073814939707517, |
| "num_tokens": 52450088.0, |
| "step": 2680 |
| }, |
| { |
| "entropy": 0.3812582534737885, |
| "epoch": 4.304, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.6246666666666668e-05, |
| "loss": 0.2996, |
| "mean_token_accuracy": 0.9058480810374021, |
| "num_tokens": 52644881.0, |
| "step": 2690 |
| }, |
| { |
| "entropy": 0.38365863654762505, |
| "epoch": 4.32, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.6224444444444446e-05, |
| "loss": 0.3067, |
| "mean_token_accuracy": 0.9044234286993742, |
| "num_tokens": 52842484.0, |
| "step": 2700 |
| }, |
| { |
| "epoch": 4.32, |
| "eval_biology_entropy": 0.5477553730010987, |
| "eval_biology_loss": 1.7485437393188477, |
| "eval_biology_mean_token_accuracy": 0.6887955894470215, |
| "eval_biology_num_tokens": 52842484.0, |
| "eval_biology_runtime": 39.2324, |
| "eval_biology_samples_per_second": 12.745, |
| "eval_biology_steps_per_second": 3.186, |
| "step": 2700 |
| }, |
| { |
| "epoch": 4.32, |
| "eval_chemistry_entropy": 0.3365655550956726, |
| "eval_chemistry_loss": 1.0836002826690674, |
| "eval_chemistry_mean_token_accuracy": 0.7849969696998597, |
| "eval_chemistry_num_tokens": 52842484.0, |
| "eval_chemistry_runtime": 48.7895, |
| "eval_chemistry_samples_per_second": 10.248, |
| "eval_chemistry_steps_per_second": 2.562, |
| "step": 2700 |
| }, |
| { |
| "entropy": 0.3800164981745183, |
| "epoch": 4.336, |
| "grad_norm": 4.25, |
| "learning_rate": 1.6202222222222223e-05, |
| "loss": 0.3048, |
| "mean_token_accuracy": 0.9049365233629942, |
| "num_tokens": 53042078.0, |
| "step": 2710 |
| }, |
| { |
| "entropy": 0.39018907845020295, |
| "epoch": 4.352, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.618e-05, |
| "loss": 0.3075, |
| "mean_token_accuracy": 0.9037735242396593, |
| "num_tokens": 53237421.0, |
| "step": 2720 |
| }, |
| { |
| "entropy": 0.385921404324472, |
| "epoch": 4.368, |
| "grad_norm": 4.375, |
| "learning_rate": 1.615777777777778e-05, |
| "loss": 0.3043, |
| "mean_token_accuracy": 0.9023260220885276, |
| "num_tokens": 53433200.0, |
| "step": 2730 |
| }, |
| { |
| "entropy": 0.39520229697227477, |
| "epoch": 4.384, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.6135555555555556e-05, |
| "loss": 0.3086, |
| "mean_token_accuracy": 0.9015748716890812, |
| "num_tokens": 53623177.0, |
| "step": 2740 |
| }, |
| { |
| "entropy": 0.3976865070872009, |
| "epoch": 4.4, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.6113333333333334e-05, |
| "loss": 0.3112, |
| "mean_token_accuracy": 0.9001961696892977, |
| "num_tokens": 53816249.0, |
| "step": 2750 |
| }, |
| { |
| "entropy": 0.3782092853449285, |
| "epoch": 4.416, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.609111111111111e-05, |
| "loss": 0.2964, |
| "mean_token_accuracy": 0.9050134517252445, |
| "num_tokens": 54019815.0, |
| "step": 2760 |
| }, |
| { |
| "entropy": 0.38737918976694347, |
| "epoch": 4.432, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.606888888888889e-05, |
| "loss": 0.3027, |
| "mean_token_accuracy": 0.9038452822715044, |
| "num_tokens": 54220035.0, |
| "step": 2770 |
| }, |
| { |
| "entropy": 0.37971124444156884, |
| "epoch": 4.448, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.6046666666666667e-05, |
| "loss": 0.3047, |
| "mean_token_accuracy": 0.9053887810558081, |
| "num_tokens": 54413944.0, |
| "step": 2780 |
| }, |
| { |
| "entropy": 0.3853255074471235, |
| "epoch": 4.464, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.6024444444444444e-05, |
| "loss": 0.3076, |
| "mean_token_accuracy": 0.9037058696150779, |
| "num_tokens": 54606029.0, |
| "step": 2790 |
| }, |
| { |
| "entropy": 0.39501711539924145, |
| "epoch": 4.48, |
| "grad_norm": 4.0, |
| "learning_rate": 1.6002222222222222e-05, |
| "loss": 0.3179, |
| "mean_token_accuracy": 0.9002406939864158, |
| "num_tokens": 54798127.0, |
| "step": 2800 |
| }, |
| { |
| "epoch": 4.48, |
| "eval_biology_entropy": 0.5481105494499207, |
| "eval_biology_loss": 1.749912142753601, |
| "eval_biology_mean_token_accuracy": 0.688213824748993, |
| "eval_biology_num_tokens": 54798127.0, |
| "eval_biology_runtime": 39.1042, |
| "eval_biology_samples_per_second": 12.786, |
| "eval_biology_steps_per_second": 3.197, |
| "step": 2800 |
| }, |
| { |
| "epoch": 4.48, |
| "eval_chemistry_entropy": 0.33537248528003694, |
| "eval_chemistry_loss": 1.081470251083374, |
| "eval_chemistry_mean_token_accuracy": 0.7852825150489807, |
| "eval_chemistry_num_tokens": 54798127.0, |
| "eval_chemistry_runtime": 48.7686, |
| "eval_chemistry_samples_per_second": 10.253, |
| "eval_chemistry_steps_per_second": 2.563, |
| "step": 2800 |
| }, |
| { |
| "entropy": 0.3815502510406077, |
| "epoch": 4.496, |
| "grad_norm": 4.0, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 0.304, |
| "mean_token_accuracy": 0.9038764249533415, |
| "num_tokens": 54997752.0, |
| "step": 2810 |
| }, |
| { |
| "entropy": 0.39848967771977184, |
| "epoch": 4.5120000000000005, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.5957777777777777e-05, |
| "loss": 0.3177, |
| "mean_token_accuracy": 0.8992222622036934, |
| "num_tokens": 55196508.0, |
| "step": 2820 |
| }, |
| { |
| "entropy": 0.3797444535419345, |
| "epoch": 4.5280000000000005, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.5935555555555558e-05, |
| "loss": 0.3029, |
| "mean_token_accuracy": 0.9034993767738342, |
| "num_tokens": 55399257.0, |
| "step": 2830 |
| }, |
| { |
| "entropy": 0.3818223585374653, |
| "epoch": 4.5440000000000005, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.5913333333333332e-05, |
| "loss": 0.3031, |
| "mean_token_accuracy": 0.9035118374973535, |
| "num_tokens": 55593334.0, |
| "step": 2840 |
| }, |
| { |
| "entropy": 0.38299790266901257, |
| "epoch": 4.5600000000000005, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.5891111111111113e-05, |
| "loss": 0.3033, |
| "mean_token_accuracy": 0.9039735317230224, |
| "num_tokens": 55789260.0, |
| "step": 2850 |
| }, |
| { |
| "entropy": 0.38522821124643086, |
| "epoch": 4.576, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.5868888888888888e-05, |
| "loss": 0.3049, |
| "mean_token_accuracy": 0.9029418785125017, |
| "num_tokens": 55983747.0, |
| "step": 2860 |
| }, |
| { |
| "entropy": 0.3893515238538384, |
| "epoch": 4.592, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.584666666666667e-05, |
| "loss": 0.3082, |
| "mean_token_accuracy": 0.900501973927021, |
| "num_tokens": 56180293.0, |
| "step": 2870 |
| }, |
| { |
| "entropy": 0.4013020605780184, |
| "epoch": 4.608, |
| "grad_norm": 4.75, |
| "learning_rate": 1.5824444444444446e-05, |
| "loss": 0.3184, |
| "mean_token_accuracy": 0.8998847845941782, |
| "num_tokens": 56366523.0, |
| "step": 2880 |
| }, |
| { |
| "entropy": 0.3970656843855977, |
| "epoch": 4.624, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.5802222222222224e-05, |
| "loss": 0.3158, |
| "mean_token_accuracy": 0.900638486072421, |
| "num_tokens": 56562990.0, |
| "step": 2890 |
| }, |
| { |
| "entropy": 0.3931910445913672, |
| "epoch": 4.64, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.578e-05, |
| "loss": 0.3142, |
| "mean_token_accuracy": 0.9009704478085041, |
| "num_tokens": 56751885.0, |
| "step": 2900 |
| }, |
| { |
| "epoch": 4.64, |
| "eval_biology_entropy": 0.5423888788223267, |
| "eval_biology_loss": 1.7526012659072876, |
| "eval_biology_mean_token_accuracy": 0.6883704071044922, |
| "eval_biology_num_tokens": 56751885.0, |
| "eval_biology_runtime": 39.1144, |
| "eval_biology_samples_per_second": 12.783, |
| "eval_biology_steps_per_second": 3.196, |
| "step": 2900 |
| }, |
| { |
| "epoch": 4.64, |
| "eval_chemistry_entropy": 0.33322553062438964, |
| "eval_chemistry_loss": 1.0912866592407227, |
| "eval_chemistry_mean_token_accuracy": 0.7841868152618409, |
| "eval_chemistry_num_tokens": 56751885.0, |
| "eval_chemistry_runtime": 48.6802, |
| "eval_chemistry_samples_per_second": 10.271, |
| "eval_chemistry_steps_per_second": 2.568, |
| "step": 2900 |
| }, |
| { |
| "entropy": 0.3936818397603929, |
| "epoch": 4.656, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.575777777777778e-05, |
| "loss": 0.317, |
| "mean_token_accuracy": 0.900015426427126, |
| "num_tokens": 56945901.0, |
| "step": 2910 |
| }, |
| { |
| "entropy": 0.3930096280761063, |
| "epoch": 4.672, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.5735555555555557e-05, |
| "loss": 0.3139, |
| "mean_token_accuracy": 0.8998955756425857, |
| "num_tokens": 57137042.0, |
| "step": 2920 |
| }, |
| { |
| "entropy": 0.39152887677773834, |
| "epoch": 4.688, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.5713333333333334e-05, |
| "loss": 0.3127, |
| "mean_token_accuracy": 0.9004812855273485, |
| "num_tokens": 57339034.0, |
| "step": 2930 |
| }, |
| { |
| "entropy": 0.3969948007725179, |
| "epoch": 4.704, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.5691111111111112e-05, |
| "loss": 0.3181, |
| "mean_token_accuracy": 0.9000732693821192, |
| "num_tokens": 57533963.0, |
| "step": 2940 |
| }, |
| { |
| "entropy": 0.40082706985995176, |
| "epoch": 4.72, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.5668888888888893e-05, |
| "loss": 0.3208, |
| "mean_token_accuracy": 0.8987932920455932, |
| "num_tokens": 57725571.0, |
| "step": 2950 |
| }, |
| { |
| "entropy": 0.38801047690212725, |
| "epoch": 4.736, |
| "grad_norm": 4.125, |
| "learning_rate": 1.5646666666666667e-05, |
| "loss": 0.3151, |
| "mean_token_accuracy": 0.9012716162949801, |
| "num_tokens": 57929053.0, |
| "step": 2960 |
| }, |
| { |
| "entropy": 0.39129406129941346, |
| "epoch": 4.752, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.5624444444444448e-05, |
| "loss": 0.3131, |
| "mean_token_accuracy": 0.9008307009935379, |
| "num_tokens": 58122816.0, |
| "step": 2970 |
| }, |
| { |
| "entropy": 0.3819296005181968, |
| "epoch": 4.768, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.5602222222222222e-05, |
| "loss": 0.3096, |
| "mean_token_accuracy": 0.9024453420192003, |
| "num_tokens": 58321778.0, |
| "step": 2980 |
| }, |
| { |
| "entropy": 0.3890405807644129, |
| "epoch": 4.784, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 0.3085, |
| "mean_token_accuracy": 0.9007145997136832, |
| "num_tokens": 58522511.0, |
| "step": 2990 |
| }, |
| { |
| "entropy": 0.3927446133457124, |
| "epoch": 4.8, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.5557777777777778e-05, |
| "loss": 0.316, |
| "mean_token_accuracy": 0.8994843460619449, |
| "num_tokens": 58712773.0, |
| "step": 3000 |
| }, |
| { |
| "epoch": 4.8, |
| "eval_biology_entropy": 0.5437517564296722, |
| "eval_biology_loss": 1.7587285041809082, |
| "eval_biology_mean_token_accuracy": 0.6887696690559387, |
| "eval_biology_num_tokens": 58712773.0, |
| "eval_biology_runtime": 39.143, |
| "eval_biology_samples_per_second": 12.774, |
| "eval_biology_steps_per_second": 3.193, |
| "step": 3000 |
| }, |
| { |
| "epoch": 4.8, |
| "eval_chemistry_entropy": 0.33195315623283383, |
| "eval_chemistry_loss": 1.0940355062484741, |
| "eval_chemistry_mean_token_accuracy": 0.784522089958191, |
| "eval_chemistry_num_tokens": 58712773.0, |
| "eval_chemistry_runtime": 48.5363, |
| "eval_chemistry_samples_per_second": 10.302, |
| "eval_chemistry_steps_per_second": 2.575, |
| "step": 3000 |
| }, |
| { |
| "entropy": 0.3984108787029982, |
| "epoch": 4.816, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.553555555555556e-05, |
| "loss": 0.313, |
| "mean_token_accuracy": 0.9002299178391695, |
| "num_tokens": 58906776.0, |
| "step": 3010 |
| }, |
| { |
| "entropy": 0.3900939176790416, |
| "epoch": 4.832, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.5513333333333333e-05, |
| "loss": 0.3127, |
| "mean_token_accuracy": 0.902032233774662, |
| "num_tokens": 59094439.0, |
| "step": 3020 |
| }, |
| { |
| "entropy": 0.4013954695314169, |
| "epoch": 4.848, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.5491111111111114e-05, |
| "loss": 0.3218, |
| "mean_token_accuracy": 0.8968057397753, |
| "num_tokens": 59283668.0, |
| "step": 3030 |
| }, |
| { |
| "entropy": 0.4007334131747484, |
| "epoch": 4.864, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.546888888888889e-05, |
| "loss": 0.3185, |
| "mean_token_accuracy": 0.8989983338862657, |
| "num_tokens": 59482325.0, |
| "step": 3040 |
| }, |
| { |
| "entropy": 0.3982712904922664, |
| "epoch": 4.88, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.544666666666667e-05, |
| "loss": 0.319, |
| "mean_token_accuracy": 0.8976643521338701, |
| "num_tokens": 59679079.0, |
| "step": 3050 |
| }, |
| { |
| "entropy": 0.393554334808141, |
| "epoch": 4.896, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.5424444444444447e-05, |
| "loss": 0.3132, |
| "mean_token_accuracy": 0.8996120911091566, |
| "num_tokens": 59878689.0, |
| "step": 3060 |
| }, |
| { |
| "entropy": 0.3952883956953883, |
| "epoch": 4.912, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.5402222222222224e-05, |
| "loss": 0.3141, |
| "mean_token_accuracy": 0.8999388955533505, |
| "num_tokens": 60073168.0, |
| "step": 3070 |
| }, |
| { |
| "entropy": 0.38866265024989843, |
| "epoch": 4.928, |
| "grad_norm": 4.125, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 0.3128, |
| "mean_token_accuracy": 0.9021519985049963, |
| "num_tokens": 60265395.0, |
| "step": 3080 |
| }, |
| { |
| "entropy": 0.3970038187690079, |
| "epoch": 4.944, |
| "grad_norm": 4.75, |
| "learning_rate": 1.535777777777778e-05, |
| "loss": 0.3174, |
| "mean_token_accuracy": 0.8994665116071701, |
| "num_tokens": 60457653.0, |
| "step": 3090 |
| }, |
| { |
| "entropy": 0.39560983879491685, |
| "epoch": 4.96, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.5335555555555557e-05, |
| "loss": 0.3142, |
| "mean_token_accuracy": 0.8998933300375939, |
| "num_tokens": 60648043.0, |
| "step": 3100 |
| }, |
| { |
| "epoch": 4.96, |
| "eval_biology_entropy": 0.5469261469841004, |
| "eval_biology_loss": 1.749172568321228, |
| "eval_biology_mean_token_accuracy": 0.6890067672729492, |
| "eval_biology_num_tokens": 60648043.0, |
| "eval_biology_runtime": 39.0921, |
| "eval_biology_samples_per_second": 12.79, |
| "eval_biology_steps_per_second": 3.198, |
| "step": 3100 |
| }, |
| { |
| "epoch": 4.96, |
| "eval_chemistry_entropy": 0.3342595580816269, |
| "eval_chemistry_loss": 1.088944911956787, |
| "eval_chemistry_mean_token_accuracy": 0.7847078347206116, |
| "eval_chemistry_num_tokens": 60648043.0, |
| "eval_chemistry_runtime": 48.6989, |
| "eval_chemistry_samples_per_second": 10.267, |
| "eval_chemistry_steps_per_second": 2.567, |
| "step": 3100 |
| }, |
| { |
| "entropy": 0.3956923006102443, |
| "epoch": 4.976, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.5313333333333335e-05, |
| "loss": 0.3168, |
| "mean_token_accuracy": 0.8992619428783655, |
| "num_tokens": 60846730.0, |
| "step": 3110 |
| }, |
| { |
| "entropy": 0.40068443985655905, |
| "epoch": 4.992, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.5291111111111112e-05, |
| "loss": 0.3234, |
| "mean_token_accuracy": 0.8976013146340847, |
| "num_tokens": 61037198.0, |
| "step": 3120 |
| }, |
| { |
| "entropy": 0.3389358670450747, |
| "epoch": 5.008, |
| "grad_norm": 3.375, |
| "learning_rate": 1.526888888888889e-05, |
| "loss": 0.2333, |
| "mean_token_accuracy": 0.9301781207323074, |
| "num_tokens": 61241298.0, |
| "step": 3130 |
| }, |
| { |
| "entropy": 0.2480829765088856, |
| "epoch": 5.024, |
| "grad_norm": 3.171875, |
| "learning_rate": 1.5246666666666668e-05, |
| "loss": 0.1434, |
| "mean_token_accuracy": 0.9616116534918546, |
| "num_tokens": 61438801.0, |
| "step": 3140 |
| }, |
| { |
| "entropy": 0.24661246789619326, |
| "epoch": 5.04, |
| "grad_norm": 2.84375, |
| "learning_rate": 1.5224444444444447e-05, |
| "loss": 0.1406, |
| "mean_token_accuracy": 0.9622246209532023, |
| "num_tokens": 61632164.0, |
| "step": 3150 |
| }, |
| { |
| "entropy": 0.23877854393795134, |
| "epoch": 5.056, |
| "grad_norm": 2.703125, |
| "learning_rate": 1.5202222222222223e-05, |
| "loss": 0.1371, |
| "mean_token_accuracy": 0.9626986987888813, |
| "num_tokens": 61825554.0, |
| "step": 3160 |
| }, |
| { |
| "entropy": 0.23843114031478763, |
| "epoch": 5.072, |
| "grad_norm": 3.171875, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 0.1363, |
| "mean_token_accuracy": 0.9628897160291672, |
| "num_tokens": 62030317.0, |
| "step": 3170 |
| }, |
| { |
| "entropy": 0.24298696424812077, |
| "epoch": 5.088, |
| "grad_norm": 2.75, |
| "learning_rate": 1.5157777777777778e-05, |
| "loss": 0.1412, |
| "mean_token_accuracy": 0.9619745567440987, |
| "num_tokens": 62223614.0, |
| "step": 3180 |
| }, |
| { |
| "entropy": 0.2342766781337559, |
| "epoch": 5.104, |
| "grad_norm": 2.703125, |
| "learning_rate": 1.5135555555555557e-05, |
| "loss": 0.1371, |
| "mean_token_accuracy": 0.9634473901242018, |
| "num_tokens": 62423057.0, |
| "step": 3190 |
| }, |
| { |
| "entropy": 0.24425409492105246, |
| "epoch": 5.12, |
| "grad_norm": 3.09375, |
| "learning_rate": 1.5113333333333335e-05, |
| "loss": 0.1439, |
| "mean_token_accuracy": 0.9604210957884789, |
| "num_tokens": 62617495.0, |
| "step": 3200 |
| }, |
| { |
| "epoch": 5.12, |
| "eval_biology_entropy": 0.4855248901844025, |
| "eval_biology_loss": 1.8822486400604248, |
| "eval_biology_mean_token_accuracy": 0.6859995684623719, |
| "eval_biology_num_tokens": 62617495.0, |
| "eval_biology_runtime": 41.7538, |
| "eval_biology_samples_per_second": 11.975, |
| "eval_biology_steps_per_second": 2.994, |
| "step": 3200 |
| }, |
| { |
| "epoch": 5.12, |
| "eval_chemistry_entropy": 0.30270042550563814, |
| "eval_chemistry_loss": 1.1746970415115356, |
| "eval_chemistry_mean_token_accuracy": 0.7822765312194824, |
| "eval_chemistry_num_tokens": 62617495.0, |
| "eval_chemistry_runtime": 50.6001, |
| "eval_chemistry_samples_per_second": 9.881, |
| "eval_chemistry_steps_per_second": 2.47, |
| "step": 3200 |
| }, |
| { |
| "entropy": 0.2399349079467356, |
| "epoch": 5.136, |
| "grad_norm": 2.984375, |
| "learning_rate": 1.5091111111111113e-05, |
| "loss": 0.1421, |
| "mean_token_accuracy": 0.9612967647612095, |
| "num_tokens": 62811652.0, |
| "step": 3210 |
| }, |
| { |
| "entropy": 0.2444787840358913, |
| "epoch": 5.152, |
| "grad_norm": 3.140625, |
| "learning_rate": 1.506888888888889e-05, |
| "loss": 0.1443, |
| "mean_token_accuracy": 0.9609584499150514, |
| "num_tokens": 63011827.0, |
| "step": 3220 |
| }, |
| { |
| "entropy": 0.24030484762042761, |
| "epoch": 5.168, |
| "grad_norm": 2.953125, |
| "learning_rate": 1.5046666666666668e-05, |
| "loss": 0.1421, |
| "mean_token_accuracy": 0.9606363896280528, |
| "num_tokens": 63204783.0, |
| "step": 3230 |
| }, |
| { |
| "entropy": 0.2405925899744034, |
| "epoch": 5.184, |
| "grad_norm": 2.796875, |
| "learning_rate": 1.5024444444444445e-05, |
| "loss": 0.1426, |
| "mean_token_accuracy": 0.9611560478806496, |
| "num_tokens": 63395495.0, |
| "step": 3240 |
| }, |
| { |
| "entropy": 0.24630163582041859, |
| "epoch": 5.2, |
| "grad_norm": 3.0, |
| "learning_rate": 1.5002222222222223e-05, |
| "loss": 0.1491, |
| "mean_token_accuracy": 0.9594012718647719, |
| "num_tokens": 63583947.0, |
| "step": 3250 |
| }, |
| { |
| "entropy": 0.23497009510174394, |
| "epoch": 5.216, |
| "grad_norm": 3.125, |
| "learning_rate": 1.498e-05, |
| "loss": 0.1418, |
| "mean_token_accuracy": 0.9615683332085609, |
| "num_tokens": 63788497.0, |
| "step": 3260 |
| }, |
| { |
| "entropy": 0.23729531140998006, |
| "epoch": 5.232, |
| "grad_norm": 3.09375, |
| "learning_rate": 1.495777777777778e-05, |
| "loss": 0.1417, |
| "mean_token_accuracy": 0.961542560532689, |
| "num_tokens": 63984458.0, |
| "step": 3270 |
| }, |
| { |
| "entropy": 0.24504502927884458, |
| "epoch": 5.248, |
| "grad_norm": 3.265625, |
| "learning_rate": 1.4935555555555556e-05, |
| "loss": 0.1487, |
| "mean_token_accuracy": 0.9591677300632, |
| "num_tokens": 64181515.0, |
| "step": 3280 |
| }, |
| { |
| "entropy": 0.24411029471084475, |
| "epoch": 5.264, |
| "grad_norm": 3.234375, |
| "learning_rate": 1.4913333333333335e-05, |
| "loss": 0.1468, |
| "mean_token_accuracy": 0.9606987874954939, |
| "num_tokens": 64371068.0, |
| "step": 3290 |
| }, |
| { |
| "entropy": 0.24654189068824053, |
| "epoch": 5.28, |
| "grad_norm": 3.1875, |
| "learning_rate": 1.4891111111111111e-05, |
| "loss": 0.1483, |
| "mean_token_accuracy": 0.9585763823240996, |
| "num_tokens": 64558621.0, |
| "step": 3300 |
| }, |
| { |
| "epoch": 5.28, |
| "eval_biology_entropy": 0.47723662209510803, |
| "eval_biology_loss": 1.8960919380187988, |
| "eval_biology_mean_token_accuracy": 0.6864940290451049, |
| "eval_biology_num_tokens": 64558621.0, |
| "eval_biology_runtime": 39.1948, |
| "eval_biology_samples_per_second": 12.757, |
| "eval_biology_steps_per_second": 3.189, |
| "step": 3300 |
| }, |
| { |
| "epoch": 5.28, |
| "eval_chemistry_entropy": 0.29984386241436006, |
| "eval_chemistry_loss": 1.1824941635131836, |
| "eval_chemistry_mean_token_accuracy": 0.7820545530319214, |
| "eval_chemistry_num_tokens": 64558621.0, |
| "eval_chemistry_runtime": 48.8041, |
| "eval_chemistry_samples_per_second": 10.245, |
| "eval_chemistry_steps_per_second": 2.561, |
| "step": 3300 |
| }, |
| { |
| "entropy": 0.2373804060742259, |
| "epoch": 5.296, |
| "grad_norm": 3.390625, |
| "learning_rate": 1.486888888888889e-05, |
| "loss": 0.1434, |
| "mean_token_accuracy": 0.9604081165045499, |
| "num_tokens": 64758641.0, |
| "step": 3310 |
| }, |
| { |
| "entropy": 0.23958202386274935, |
| "epoch": 5.312, |
| "grad_norm": 2.890625, |
| "learning_rate": 1.4846666666666666e-05, |
| "loss": 0.143, |
| "mean_token_accuracy": 0.9606925457715988, |
| "num_tokens": 64956086.0, |
| "step": 3320 |
| }, |
| { |
| "entropy": 0.24059962034225463, |
| "epoch": 5.328, |
| "grad_norm": 3.359375, |
| "learning_rate": 1.4824444444444446e-05, |
| "loss": 0.1475, |
| "mean_token_accuracy": 0.9589107587933541, |
| "num_tokens": 65156894.0, |
| "step": 3330 |
| }, |
| { |
| "entropy": 0.2476472335867584, |
| "epoch": 5.344, |
| "grad_norm": 3.0625, |
| "learning_rate": 1.4802222222222225e-05, |
| "loss": 0.1491, |
| "mean_token_accuracy": 0.9589390732347965, |
| "num_tokens": 65345813.0, |
| "step": 3340 |
| }, |
| { |
| "entropy": 0.24589035334065557, |
| "epoch": 5.36, |
| "grad_norm": 3.09375, |
| "learning_rate": 1.478e-05, |
| "loss": 0.1471, |
| "mean_token_accuracy": 0.9593054607510567, |
| "num_tokens": 65534871.0, |
| "step": 3350 |
| }, |
| { |
| "entropy": 0.24193211700767278, |
| "epoch": 5.376, |
| "grad_norm": 3.34375, |
| "learning_rate": 1.475777777777778e-05, |
| "loss": 0.148, |
| "mean_token_accuracy": 0.9590138528496027, |
| "num_tokens": 65732993.0, |
| "step": 3360 |
| }, |
| { |
| "entropy": 0.24610914513468743, |
| "epoch": 5.392, |
| "grad_norm": 2.9375, |
| "learning_rate": 1.4735555555555556e-05, |
| "loss": 0.1486, |
| "mean_token_accuracy": 0.9580031178891659, |
| "num_tokens": 65934902.0, |
| "step": 3370 |
| }, |
| { |
| "entropy": 0.24404912833124398, |
| "epoch": 5.408, |
| "grad_norm": 3.28125, |
| "learning_rate": 1.4713333333333335e-05, |
| "loss": 0.1499, |
| "mean_token_accuracy": 0.958684965595603, |
| "num_tokens": 66125726.0, |
| "step": 3380 |
| }, |
| { |
| "entropy": 0.2426779827103019, |
| "epoch": 5.424, |
| "grad_norm": 3.234375, |
| "learning_rate": 1.4691111111111111e-05, |
| "loss": 0.1489, |
| "mean_token_accuracy": 0.9597231402993203, |
| "num_tokens": 66323902.0, |
| "step": 3390 |
| }, |
| { |
| "entropy": 0.24080183450132608, |
| "epoch": 5.44, |
| "grad_norm": 3.546875, |
| "learning_rate": 1.466888888888889e-05, |
| "loss": 0.1492, |
| "mean_token_accuracy": 0.9590619742870331, |
| "num_tokens": 66512056.0, |
| "step": 3400 |
| }, |
| { |
| "epoch": 5.44, |
| "eval_biology_entropy": 0.474915821313858, |
| "eval_biology_loss": 1.9035038948059082, |
| "eval_biology_mean_token_accuracy": 0.6857102961540222, |
| "eval_biology_num_tokens": 66512056.0, |
| "eval_biology_runtime": 39.0895, |
| "eval_biology_samples_per_second": 12.791, |
| "eval_biology_steps_per_second": 3.198, |
| "step": 3400 |
| }, |
| { |
| "epoch": 5.44, |
| "eval_chemistry_entropy": 0.2978140150308609, |
| "eval_chemistry_loss": 1.1872378587722778, |
| "eval_chemistry_mean_token_accuracy": 0.7817945637702942, |
| "eval_chemistry_num_tokens": 66512056.0, |
| "eval_chemistry_runtime": 48.7007, |
| "eval_chemistry_samples_per_second": 10.267, |
| "eval_chemistry_steps_per_second": 2.567, |
| "step": 3400 |
| }, |
| { |
| "entropy": 0.24143663756549358, |
| "epoch": 5.456, |
| "grad_norm": 2.765625, |
| "learning_rate": 1.4646666666666666e-05, |
| "loss": 0.1474, |
| "mean_token_accuracy": 0.9594498135149478, |
| "num_tokens": 66709891.0, |
| "step": 3410 |
| }, |
| { |
| "entropy": 0.24799445318058133, |
| "epoch": 5.4719999999999995, |
| "grad_norm": 3.265625, |
| "learning_rate": 1.4624444444444446e-05, |
| "loss": 0.1501, |
| "mean_token_accuracy": 0.9580833919346332, |
| "num_tokens": 66902562.0, |
| "step": 3420 |
| }, |
| { |
| "entropy": 0.247726259380579, |
| "epoch": 5.4879999999999995, |
| "grad_norm": 2.90625, |
| "learning_rate": 1.4602222222222225e-05, |
| "loss": 0.1538, |
| "mean_token_accuracy": 0.9573916871100664, |
| "num_tokens": 67094381.0, |
| "step": 3430 |
| }, |
| { |
| "entropy": 0.24004812752828003, |
| "epoch": 5.504, |
| "grad_norm": 3.359375, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 0.1477, |
| "mean_token_accuracy": 0.9594778578728438, |
| "num_tokens": 67289859.0, |
| "step": 3440 |
| }, |
| { |
| "entropy": 0.25117804668843746, |
| "epoch": 5.52, |
| "grad_norm": 3.296875, |
| "learning_rate": 1.455777777777778e-05, |
| "loss": 0.1544, |
| "mean_token_accuracy": 0.9566788725554943, |
| "num_tokens": 67476884.0, |
| "step": 3450 |
| }, |
| { |
| "entropy": 0.24327809335663914, |
| "epoch": 5.536, |
| "grad_norm": 3.25, |
| "learning_rate": 1.4535555555555556e-05, |
| "loss": 0.1494, |
| "mean_token_accuracy": 0.958500000461936, |
| "num_tokens": 67670050.0, |
| "step": 3460 |
| }, |
| { |
| "entropy": 0.2430077507160604, |
| "epoch": 5.552, |
| "grad_norm": 3.15625, |
| "learning_rate": 1.4513333333333336e-05, |
| "loss": 0.1504, |
| "mean_token_accuracy": 0.9591543607413768, |
| "num_tokens": 67863965.0, |
| "step": 3470 |
| }, |
| { |
| "entropy": 0.23817884922027588, |
| "epoch": 5.568, |
| "grad_norm": 2.65625, |
| "learning_rate": 1.4491111111111111e-05, |
| "loss": 0.1474, |
| "mean_token_accuracy": 0.9588000290095806, |
| "num_tokens": 68065585.0, |
| "step": 3480 |
| }, |
| { |
| "entropy": 0.23812123024836182, |
| "epoch": 5.584, |
| "grad_norm": 2.84375, |
| "learning_rate": 1.446888888888889e-05, |
| "loss": 0.1469, |
| "mean_token_accuracy": 0.959443487599492, |
| "num_tokens": 68265404.0, |
| "step": 3490 |
| }, |
| { |
| "entropy": 0.24377540815621615, |
| "epoch": 5.6, |
| "grad_norm": 3.265625, |
| "learning_rate": 1.4446666666666668e-05, |
| "loss": 0.1485, |
| "mean_token_accuracy": 0.9580644488334655, |
| "num_tokens": 68461451.0, |
| "step": 3500 |
| }, |
| { |
| "epoch": 5.6, |
| "eval_biology_entropy": 0.4784146144390106, |
| "eval_biology_loss": 1.9023874998092651, |
| "eval_biology_mean_token_accuracy": 0.6854758291244507, |
| "eval_biology_num_tokens": 68461451.0, |
| "eval_biology_runtime": 39.0892, |
| "eval_biology_samples_per_second": 12.791, |
| "eval_biology_steps_per_second": 3.198, |
| "step": 3500 |
| }, |
| { |
| "epoch": 5.6, |
| "eval_chemistry_entropy": 0.29993112993240356, |
| "eval_chemistry_loss": 1.1889783143997192, |
| "eval_chemistry_mean_token_accuracy": 0.781288462638855, |
| "eval_chemistry_num_tokens": 68461451.0, |
| "eval_chemistry_runtime": 48.5405, |
| "eval_chemistry_samples_per_second": 10.301, |
| "eval_chemistry_steps_per_second": 2.575, |
| "step": 3500 |
| }, |
| { |
| "entropy": 0.23950189482420683, |
| "epoch": 5.616, |
| "grad_norm": 3.125, |
| "learning_rate": 1.4424444444444446e-05, |
| "loss": 0.1475, |
| "mean_token_accuracy": 0.9595385141670704, |
| "num_tokens": 68660370.0, |
| "step": 3510 |
| }, |
| { |
| "entropy": 0.24690509317442774, |
| "epoch": 5.632, |
| "grad_norm": 3.296875, |
| "learning_rate": 1.4402222222222224e-05, |
| "loss": 0.1526, |
| "mean_token_accuracy": 0.9578721337020397, |
| "num_tokens": 68853923.0, |
| "step": 3520 |
| }, |
| { |
| "entropy": 0.2501044309698045, |
| "epoch": 5.648, |
| "grad_norm": 2.984375, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 0.1559, |
| "mean_token_accuracy": 0.9569032177329063, |
| "num_tokens": 69043849.0, |
| "step": 3530 |
| }, |
| { |
| "entropy": 0.24690743535757065, |
| "epoch": 5.664, |
| "grad_norm": 2.921875, |
| "learning_rate": 1.4357777777777779e-05, |
| "loss": 0.1515, |
| "mean_token_accuracy": 0.9578557062894106, |
| "num_tokens": 69239790.0, |
| "step": 3540 |
| }, |
| { |
| "entropy": 0.2420658844523132, |
| "epoch": 5.68, |
| "grad_norm": 3.25, |
| "learning_rate": 1.4335555555555556e-05, |
| "loss": 0.1496, |
| "mean_token_accuracy": 0.9580045863986015, |
| "num_tokens": 69440632.0, |
| "step": 3550 |
| }, |
| { |
| "entropy": 0.2461313429288566, |
| "epoch": 5.696, |
| "grad_norm": 3.21875, |
| "learning_rate": 1.4313333333333334e-05, |
| "loss": 0.1529, |
| "mean_token_accuracy": 0.9572788786143065, |
| "num_tokens": 69639648.0, |
| "step": 3560 |
| }, |
| { |
| "entropy": 0.24486252525821328, |
| "epoch": 5.712, |
| "grad_norm": 3.28125, |
| "learning_rate": 1.4291111111111113e-05, |
| "loss": 0.1509, |
| "mean_token_accuracy": 0.9584188058972358, |
| "num_tokens": 69836856.0, |
| "step": 3570 |
| }, |
| { |
| "entropy": 0.23996384926140307, |
| "epoch": 5.728, |
| "grad_norm": 3.484375, |
| "learning_rate": 1.426888888888889e-05, |
| "loss": 0.1504, |
| "mean_token_accuracy": 0.9579487156122923, |
| "num_tokens": 70038312.0, |
| "step": 3580 |
| }, |
| { |
| "entropy": 0.24508317830041051, |
| "epoch": 5.744, |
| "grad_norm": 3.484375, |
| "learning_rate": 1.4246666666666669e-05, |
| "loss": 0.15, |
| "mean_token_accuracy": 0.9580871388316154, |
| "num_tokens": 70236131.0, |
| "step": 3590 |
| }, |
| { |
| "entropy": 0.24095194637775422, |
| "epoch": 5.76, |
| "grad_norm": 3.28125, |
| "learning_rate": 1.4224444444444445e-05, |
| "loss": 0.1493, |
| "mean_token_accuracy": 0.9590608909726143, |
| "num_tokens": 70432765.0, |
| "step": 3600 |
| }, |
| { |
| "epoch": 5.76, |
| "eval_biology_entropy": 0.47431311655044556, |
| "eval_biology_loss": 1.9089088439941406, |
| "eval_biology_mean_token_accuracy": 0.6859101238250732, |
| "eval_biology_num_tokens": 70432765.0, |
| "eval_biology_runtime": 39.1267, |
| "eval_biology_samples_per_second": 12.779, |
| "eval_biology_steps_per_second": 3.195, |
| "step": 3600 |
| }, |
| { |
| "epoch": 5.76, |
| "eval_chemistry_entropy": 0.29715424621105196, |
| "eval_chemistry_loss": 1.1934040784835815, |
| "eval_chemistry_mean_token_accuracy": 0.7814722752571106, |
| "eval_chemistry_num_tokens": 70432765.0, |
| "eval_chemistry_runtime": 48.8292, |
| "eval_chemistry_samples_per_second": 10.24, |
| "eval_chemistry_steps_per_second": 2.56, |
| "step": 3600 |
| }, |
| { |
| "entropy": 0.2489096809178591, |
| "epoch": 5.776, |
| "grad_norm": 2.953125, |
| "learning_rate": 1.4202222222222224e-05, |
| "loss": 0.1526, |
| "mean_token_accuracy": 0.9566314175724984, |
| "num_tokens": 70627443.0, |
| "step": 3610 |
| }, |
| { |
| "entropy": 0.23710308950394393, |
| "epoch": 5.792, |
| "grad_norm": 3.15625, |
| "learning_rate": 1.418e-05, |
| "loss": 0.1484, |
| "mean_token_accuracy": 0.9586101662367582, |
| "num_tokens": 70826637.0, |
| "step": 3620 |
| }, |
| { |
| "entropy": 0.24433835800737141, |
| "epoch": 5.808, |
| "grad_norm": 3.4375, |
| "learning_rate": 1.4157777777777779e-05, |
| "loss": 0.1518, |
| "mean_token_accuracy": 0.9571759428828954, |
| "num_tokens": 71023895.0, |
| "step": 3630 |
| }, |
| { |
| "entropy": 0.24292335715144872, |
| "epoch": 5.824, |
| "grad_norm": 3.25, |
| "learning_rate": 1.4135555555555555e-05, |
| "loss": 0.1508, |
| "mean_token_accuracy": 0.9575801249593496, |
| "num_tokens": 71219734.0, |
| "step": 3640 |
| }, |
| { |
| "entropy": 0.24374447409063577, |
| "epoch": 5.84, |
| "grad_norm": 3.234375, |
| "learning_rate": 1.4113333333333334e-05, |
| "loss": 0.1517, |
| "mean_token_accuracy": 0.9578015189617872, |
| "num_tokens": 71416653.0, |
| "step": 3650 |
| }, |
| { |
| "entropy": 0.24155271723866462, |
| "epoch": 5.856, |
| "grad_norm": 2.859375, |
| "learning_rate": 1.4091111111111114e-05, |
| "loss": 0.1506, |
| "mean_token_accuracy": 0.9578181304037571, |
| "num_tokens": 71612764.0, |
| "step": 3660 |
| }, |
| { |
| "entropy": 0.24872884480282664, |
| "epoch": 5.872, |
| "grad_norm": 3.46875, |
| "learning_rate": 1.406888888888889e-05, |
| "loss": 0.1523, |
| "mean_token_accuracy": 0.9566898606717587, |
| "num_tokens": 71810178.0, |
| "step": 3670 |
| }, |
| { |
| "entropy": 0.24792636511847377, |
| "epoch": 5.888, |
| "grad_norm": 3.203125, |
| "learning_rate": 1.4046666666666669e-05, |
| "loss": 0.1518, |
| "mean_token_accuracy": 0.9570049945265054, |
| "num_tokens": 72000737.0, |
| "step": 3680 |
| }, |
| { |
| "entropy": 0.24725554306060077, |
| "epoch": 5.904, |
| "grad_norm": 3.03125, |
| "learning_rate": 1.4024444444444445e-05, |
| "loss": 0.1532, |
| "mean_token_accuracy": 0.957570206001401, |
| "num_tokens": 72195665.0, |
| "step": 3690 |
| }, |
| { |
| "entropy": 0.24413193976506592, |
| "epoch": 5.92, |
| "grad_norm": 3.375, |
| "learning_rate": 1.4002222222222224e-05, |
| "loss": 0.1534, |
| "mean_token_accuracy": 0.9571807373315095, |
| "num_tokens": 72390616.0, |
| "step": 3700 |
| }, |
| { |
| "epoch": 5.92, |
| "eval_biology_entropy": 0.4760978572368622, |
| "eval_biology_loss": 1.9012237787246704, |
| "eval_biology_mean_token_accuracy": 0.6856249871253968, |
| "eval_biology_num_tokens": 72390616.0, |
| "eval_biology_runtime": 39.107, |
| "eval_biology_samples_per_second": 12.785, |
| "eval_biology_steps_per_second": 3.196, |
| "step": 3700 |
| }, |
| { |
| "epoch": 5.92, |
| "eval_chemistry_entropy": 0.2979493819475174, |
| "eval_chemistry_loss": 1.1911438703536987, |
| "eval_chemistry_mean_token_accuracy": 0.7814780764579773, |
| "eval_chemistry_num_tokens": 72390616.0, |
| "eval_chemistry_runtime": 48.7139, |
| "eval_chemistry_samples_per_second": 10.264, |
| "eval_chemistry_steps_per_second": 2.566, |
| "step": 3700 |
| }, |
| { |
| "entropy": 0.2468944384716451, |
| "epoch": 5.936, |
| "grad_norm": 3.40625, |
| "learning_rate": 1.398e-05, |
| "loss": 0.1535, |
| "mean_token_accuracy": 0.9570021212100983, |
| "num_tokens": 72585041.0, |
| "step": 3710 |
| }, |
| { |
| "entropy": 0.2445361793041229, |
| "epoch": 5.952, |
| "grad_norm": 3.484375, |
| "learning_rate": 1.395777777777778e-05, |
| "loss": 0.1516, |
| "mean_token_accuracy": 0.9581042151898146, |
| "num_tokens": 72773803.0, |
| "step": 3720 |
| }, |
| { |
| "entropy": 0.2447666016407311, |
| "epoch": 5.968, |
| "grad_norm": 3.140625, |
| "learning_rate": 1.3935555555555557e-05, |
| "loss": 0.1515, |
| "mean_token_accuracy": 0.9580264937132597, |
| "num_tokens": 72973667.0, |
| "step": 3730 |
| }, |
| { |
| "entropy": 0.24522863822057844, |
| "epoch": 5.984, |
| "grad_norm": 3.609375, |
| "learning_rate": 1.3913333333333335e-05, |
| "loss": 0.1521, |
| "mean_token_accuracy": 0.9570143446326256, |
| "num_tokens": 73168554.0, |
| "step": 3740 |
| }, |
| { |
| "entropy": 0.24672735193744302, |
| "epoch": 6.0, |
| "grad_norm": 3.484375, |
| "learning_rate": 1.3891111111111114e-05, |
| "loss": 0.1535, |
| "mean_token_accuracy": 0.9568115711212158, |
| "num_tokens": 73366632.0, |
| "step": 3750 |
| }, |
| { |
| "entropy": 0.1643500389996916, |
| "epoch": 6.016, |
| "grad_norm": 1.8515625, |
| "learning_rate": 1.386888888888889e-05, |
| "loss": 0.0794, |
| "mean_token_accuracy": 0.981878663226962, |
| "num_tokens": 73566300.0, |
| "step": 3760 |
| }, |
| { |
| "entropy": 0.15798100209794938, |
| "epoch": 6.032, |
| "grad_norm": 1.7578125, |
| "learning_rate": 1.3846666666666669e-05, |
| "loss": 0.0808, |
| "mean_token_accuracy": 0.9813016086816788, |
| "num_tokens": 73761436.0, |
| "step": 3770 |
| }, |
| { |
| "entropy": 0.1547636122442782, |
| "epoch": 6.048, |
| "grad_norm": 2.390625, |
| "learning_rate": 1.3824444444444445e-05, |
| "loss": 0.078, |
| "mean_token_accuracy": 0.9817892149090767, |
| "num_tokens": 73966976.0, |
| "step": 3780 |
| }, |
| { |
| "entropy": 0.1573083738796413, |
| "epoch": 6.064, |
| "grad_norm": 2.21875, |
| "learning_rate": 1.3802222222222224e-05, |
| "loss": 0.0808, |
| "mean_token_accuracy": 0.981355207040906, |
| "num_tokens": 74164615.0, |
| "step": 3790 |
| }, |
| { |
| "entropy": 0.15763011309318245, |
| "epoch": 6.08, |
| "grad_norm": 1.765625, |
| "learning_rate": 1.378e-05, |
| "loss": 0.0812, |
| "mean_token_accuracy": 0.9804638106375932, |
| "num_tokens": 74362133.0, |
| "step": 3800 |
| }, |
| { |
| "epoch": 6.08, |
| "eval_biology_entropy": 0.4429698960781097, |
| "eval_biology_loss": 1.9905930757522583, |
| "eval_biology_mean_token_accuracy": 0.6869594359397888, |
| "eval_biology_num_tokens": 74362133.0, |
| "eval_biology_runtime": 39.086, |
| "eval_biology_samples_per_second": 12.792, |
| "eval_biology_steps_per_second": 3.198, |
| "step": 3800 |
| }, |
| { |
| "epoch": 6.08, |
| "eval_chemistry_entropy": 0.27817737650871277, |
| "eval_chemistry_loss": 1.2527207136154175, |
| "eval_chemistry_mean_token_accuracy": 0.7819559755325317, |
| "eval_chemistry_num_tokens": 74362133.0, |
| "eval_chemistry_runtime": 48.6982, |
| "eval_chemistry_samples_per_second": 10.267, |
| "eval_chemistry_steps_per_second": 2.567, |
| "step": 3800 |
| }, |
| { |
| "entropy": 0.16140781799331308, |
| "epoch": 6.096, |
| "grad_norm": 1.953125, |
| "learning_rate": 1.375777777777778e-05, |
| "loss": 0.0826, |
| "mean_token_accuracy": 0.9801803410053254, |
| "num_tokens": 74552080.0, |
| "step": 3810 |
| }, |
| { |
| "entropy": 0.1571094446349889, |
| "epoch": 6.112, |
| "grad_norm": 1.6796875, |
| "learning_rate": 1.3735555555555557e-05, |
| "loss": 0.0802, |
| "mean_token_accuracy": 0.9812397368252277, |
| "num_tokens": 74747122.0, |
| "step": 3820 |
| }, |
| { |
| "entropy": 0.1561655880883336, |
| "epoch": 6.128, |
| "grad_norm": 1.71875, |
| "learning_rate": 1.3713333333333335e-05, |
| "loss": 0.0793, |
| "mean_token_accuracy": 0.9810388378798962, |
| "num_tokens": 74951934.0, |
| "step": 3830 |
| }, |
| { |
| "entropy": 0.16185828112065792, |
| "epoch": 6.144, |
| "grad_norm": 1.8984375, |
| "learning_rate": 1.3691111111111112e-05, |
| "loss": 0.0844, |
| "mean_token_accuracy": 0.9795913517475128, |
| "num_tokens": 75143606.0, |
| "step": 3840 |
| }, |
| { |
| "entropy": 0.1595824571326375, |
| "epoch": 6.16, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.366888888888889e-05, |
| "loss": 0.0844, |
| "mean_token_accuracy": 0.9803187143057585, |
| "num_tokens": 75338375.0, |
| "step": 3850 |
| }, |
| { |
| "entropy": 0.157256978796795, |
| "epoch": 6.176, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.3646666666666668e-05, |
| "loss": 0.0815, |
| "mean_token_accuracy": 0.980838020145893, |
| "num_tokens": 75534946.0, |
| "step": 3860 |
| }, |
| { |
| "entropy": 0.1611268393229693, |
| "epoch": 6.192, |
| "grad_norm": 2.453125, |
| "learning_rate": 1.3624444444444445e-05, |
| "loss": 0.084, |
| "mean_token_accuracy": 0.9800396807491779, |
| "num_tokens": 75726244.0, |
| "step": 3870 |
| }, |
| { |
| "entropy": 0.1604965917766094, |
| "epoch": 6.208, |
| "grad_norm": 2.171875, |
| "learning_rate": 1.3602222222222223e-05, |
| "loss": 0.0841, |
| "mean_token_accuracy": 0.9799326334148646, |
| "num_tokens": 75919000.0, |
| "step": 3880 |
| }, |
| { |
| "entropy": 0.15303698368370533, |
| "epoch": 6.224, |
| "grad_norm": 1.8515625, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 0.0802, |
| "mean_token_accuracy": 0.9814254630357027, |
| "num_tokens": 76123496.0, |
| "step": 3890 |
| }, |
| { |
| "entropy": 0.15868093222379684, |
| "epoch": 6.24, |
| "grad_norm": 1.984375, |
| "learning_rate": 1.3557777777777778e-05, |
| "loss": 0.0831, |
| "mean_token_accuracy": 0.9804290477186441, |
| "num_tokens": 76318639.0, |
| "step": 3900 |
| }, |
| { |
| "epoch": 6.24, |
| "eval_biology_entropy": 0.43731211638450623, |
| "eval_biology_loss": 2.0031447410583496, |
| "eval_biology_mean_token_accuracy": 0.6863418831825256, |
| "eval_biology_num_tokens": 76318639.0, |
| "eval_biology_runtime": 39.0826, |
| "eval_biology_samples_per_second": 12.793, |
| "eval_biology_steps_per_second": 3.198, |
| "step": 3900 |
| }, |
| { |
| "epoch": 6.24, |
| "eval_chemistry_entropy": 0.2758142249584198, |
| "eval_chemistry_loss": 1.2575671672821045, |
| "eval_chemistry_mean_token_accuracy": 0.7814622535705567, |
| "eval_chemistry_num_tokens": 76318639.0, |
| "eval_chemistry_runtime": 48.6745, |
| "eval_chemistry_samples_per_second": 10.272, |
| "eval_chemistry_steps_per_second": 2.568, |
| "step": 3900 |
| }, |
| { |
| "entropy": 0.15887791737914086, |
| "epoch": 6.256, |
| "grad_norm": 1.921875, |
| "learning_rate": 1.3535555555555557e-05, |
| "loss": 0.0835, |
| "mean_token_accuracy": 0.9800458054989576, |
| "num_tokens": 76514642.0, |
| "step": 3910 |
| }, |
| { |
| "entropy": 0.15669864052906632, |
| "epoch": 6.272, |
| "grad_norm": 1.90625, |
| "learning_rate": 1.3513333333333333e-05, |
| "loss": 0.0834, |
| "mean_token_accuracy": 0.9800447918474674, |
| "num_tokens": 76712255.0, |
| "step": 3920 |
| }, |
| { |
| "entropy": 0.15718483147211373, |
| "epoch": 6.288, |
| "grad_norm": 1.9765625, |
| "learning_rate": 1.3491111111111113e-05, |
| "loss": 0.0818, |
| "mean_token_accuracy": 0.9811201948672533, |
| "num_tokens": 76911644.0, |
| "step": 3930 |
| }, |
| { |
| "entropy": 0.15946718030609192, |
| "epoch": 6.304, |
| "grad_norm": 1.9765625, |
| "learning_rate": 1.3468888888888888e-05, |
| "loss": 0.0844, |
| "mean_token_accuracy": 0.9798561457544566, |
| "num_tokens": 77105417.0, |
| "step": 3940 |
| }, |
| { |
| "entropy": 0.15860112281516195, |
| "epoch": 6.32, |
| "grad_norm": 1.90625, |
| "learning_rate": 1.3446666666666668e-05, |
| "loss": 0.085, |
| "mean_token_accuracy": 0.979928421229124, |
| "num_tokens": 77300838.0, |
| "step": 3950 |
| }, |
| { |
| "entropy": 0.16134029026143254, |
| "epoch": 6.336, |
| "grad_norm": 1.9921875, |
| "learning_rate": 1.3424444444444447e-05, |
| "loss": 0.0847, |
| "mean_token_accuracy": 0.9798865705728531, |
| "num_tokens": 77494238.0, |
| "step": 3960 |
| }, |
| { |
| "entropy": 0.1597167048137635, |
| "epoch": 6.352, |
| "grad_norm": 1.953125, |
| "learning_rate": 1.3402222222222223e-05, |
| "loss": 0.0846, |
| "mean_token_accuracy": 0.9803135544061661, |
| "num_tokens": 77686911.0, |
| "step": 3970 |
| }, |
| { |
| "entropy": 0.15899178553372623, |
| "epoch": 6.368, |
| "grad_norm": 1.9765625, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 0.0845, |
| "mean_token_accuracy": 0.9801733657717705, |
| "num_tokens": 77885839.0, |
| "step": 3980 |
| }, |
| { |
| "entropy": 0.1599136048462242, |
| "epoch": 6.384, |
| "grad_norm": 1.7734375, |
| "learning_rate": 1.3357777777777778e-05, |
| "loss": 0.0843, |
| "mean_token_accuracy": 0.9805333554744721, |
| "num_tokens": 78080377.0, |
| "step": 3990 |
| }, |
| { |
| "entropy": 0.16270005167461932, |
| "epoch": 6.4, |
| "grad_norm": 1.9609375, |
| "learning_rate": 1.3335555555555558e-05, |
| "loss": 0.0861, |
| "mean_token_accuracy": 0.9791812360286712, |
| "num_tokens": 78272695.0, |
| "step": 4000 |
| }, |
| { |
| "epoch": 6.4, |
| "eval_biology_entropy": 0.4339142458438873, |
| "eval_biology_loss": 2.013932228088379, |
| "eval_biology_mean_token_accuracy": 0.6871605515480042, |
| "eval_biology_num_tokens": 78272695.0, |
| "eval_biology_runtime": 39.1335, |
| "eval_biology_samples_per_second": 12.777, |
| "eval_biology_steps_per_second": 3.194, |
| "step": 4000 |
| }, |
| { |
| "epoch": 6.4, |
| "eval_chemistry_entropy": 0.2722398822307587, |
| "eval_chemistry_loss": 1.264957070350647, |
| "eval_chemistry_mean_token_accuracy": 0.7815649762153626, |
| "eval_chemistry_num_tokens": 78272695.0, |
| "eval_chemistry_runtime": 48.5296, |
| "eval_chemistry_samples_per_second": 10.303, |
| "eval_chemistry_steps_per_second": 2.576, |
| "step": 4000 |
| }, |
| { |
| "entropy": 0.15887724393978714, |
| "epoch": 6.416, |
| "grad_norm": 2.140625, |
| "learning_rate": 1.3313333333333333e-05, |
| "loss": 0.0841, |
| "mean_token_accuracy": 0.979737200960517, |
| "num_tokens": 78470996.0, |
| "step": 4010 |
| }, |
| { |
| "entropy": 0.1555471509229392, |
| "epoch": 6.432, |
| "grad_norm": 1.7734375, |
| "learning_rate": 1.3291111111111113e-05, |
| "loss": 0.0819, |
| "mean_token_accuracy": 0.9807171281427145, |
| "num_tokens": 78672932.0, |
| "step": 4020 |
| }, |
| { |
| "entropy": 0.16478075762279332, |
| "epoch": 6.448, |
| "grad_norm": 2.109375, |
| "learning_rate": 1.3268888888888889e-05, |
| "loss": 0.0884, |
| "mean_token_accuracy": 0.9790490448474884, |
| "num_tokens": 78858133.0, |
| "step": 4030 |
| }, |
| { |
| "entropy": 0.1630937892012298, |
| "epoch": 6.464, |
| "grad_norm": 2.140625, |
| "learning_rate": 1.3246666666666668e-05, |
| "loss": 0.0873, |
| "mean_token_accuracy": 0.9797899898141622, |
| "num_tokens": 79047017.0, |
| "step": 4040 |
| }, |
| { |
| "entropy": 0.15713603873737156, |
| "epoch": 6.48, |
| "grad_norm": 1.78125, |
| "learning_rate": 1.3224444444444446e-05, |
| "loss": 0.0823, |
| "mean_token_accuracy": 0.9800321880728007, |
| "num_tokens": 79250529.0, |
| "step": 4050 |
| }, |
| { |
| "entropy": 0.16060890816152096, |
| "epoch": 6.496, |
| "grad_norm": 1.9453125, |
| "learning_rate": 1.3202222222222223e-05, |
| "loss": 0.0847, |
| "mean_token_accuracy": 0.9796078413724899, |
| "num_tokens": 79445533.0, |
| "step": 4060 |
| }, |
| { |
| "entropy": 0.16455780765973033, |
| "epoch": 6.5120000000000005, |
| "grad_norm": 2.0, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 0.0867, |
| "mean_token_accuracy": 0.9789722558110953, |
| "num_tokens": 79633362.0, |
| "step": 4070 |
| }, |
| { |
| "entropy": 0.15953036155551673, |
| "epoch": 6.5280000000000005, |
| "grad_norm": 1.9140625, |
| "learning_rate": 1.3157777777777778e-05, |
| "loss": 0.0862, |
| "mean_token_accuracy": 0.9796191670000554, |
| "num_tokens": 79825530.0, |
| "step": 4080 |
| }, |
| { |
| "entropy": 0.16383043727837504, |
| "epoch": 6.5440000000000005, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.3135555555555558e-05, |
| "loss": 0.0867, |
| "mean_token_accuracy": 0.9795941691845655, |
| "num_tokens": 80019408.0, |
| "step": 4090 |
| }, |
| { |
| "entropy": 0.15884415335021912, |
| "epoch": 6.5600000000000005, |
| "grad_norm": 1.8984375, |
| "learning_rate": 1.3113333333333334e-05, |
| "loss": 0.0861, |
| "mean_token_accuracy": 0.9795280620455742, |
| "num_tokens": 80215849.0, |
| "step": 4100 |
| }, |
| { |
| "epoch": 6.5600000000000005, |
| "eval_biology_entropy": 0.43253094363212585, |
| "eval_biology_loss": 2.0217366218566895, |
| "eval_biology_mean_token_accuracy": 0.6865130105018615, |
| "eval_biology_num_tokens": 80215849.0, |
| "eval_biology_runtime": 39.0559, |
| "eval_biology_samples_per_second": 12.802, |
| "eval_biology_steps_per_second": 3.201, |
| "step": 4100 |
| }, |
| { |
| "epoch": 6.5600000000000005, |
| "eval_chemistry_entropy": 0.27036730325222014, |
| "eval_chemistry_loss": 1.2734538316726685, |
| "eval_chemistry_mean_token_accuracy": 0.7815690083503724, |
| "eval_chemistry_num_tokens": 80215849.0, |
| "eval_chemistry_runtime": 48.705, |
| "eval_chemistry_samples_per_second": 10.266, |
| "eval_chemistry_steps_per_second": 2.566, |
| "step": 4100 |
| }, |
| { |
| "entropy": 0.1602792263031006, |
| "epoch": 6.576, |
| "grad_norm": 2.28125, |
| "learning_rate": 1.3091111111111113e-05, |
| "loss": 0.0853, |
| "mean_token_accuracy": 0.9798764575272798, |
| "num_tokens": 80410633.0, |
| "step": 4110 |
| }, |
| { |
| "entropy": 0.15667269434779882, |
| "epoch": 6.592, |
| "grad_norm": 1.9765625, |
| "learning_rate": 1.306888888888889e-05, |
| "loss": 0.0836, |
| "mean_token_accuracy": 0.9801121633499861, |
| "num_tokens": 80611246.0, |
| "step": 4120 |
| }, |
| { |
| "entropy": 0.1650182214099914, |
| "epoch": 6.608, |
| "grad_norm": 2.15625, |
| "learning_rate": 1.3046666666666668e-05, |
| "loss": 0.0873, |
| "mean_token_accuracy": 0.9790001824498177, |
| "num_tokens": 80797745.0, |
| "step": 4130 |
| }, |
| { |
| "entropy": 0.15939321205951273, |
| "epoch": 6.624, |
| "grad_norm": 1.8671875, |
| "learning_rate": 1.3024444444444446e-05, |
| "loss": 0.0849, |
| "mean_token_accuracy": 0.9796331495046615, |
| "num_tokens": 80995217.0, |
| "step": 4140 |
| }, |
| { |
| "entropy": 0.15765618700534106, |
| "epoch": 6.64, |
| "grad_norm": 2.28125, |
| "learning_rate": 1.3002222222222223e-05, |
| "loss": 0.0844, |
| "mean_token_accuracy": 0.9795310266315937, |
| "num_tokens": 81193529.0, |
| "step": 4150 |
| }, |
| { |
| "entropy": 0.1607879212126136, |
| "epoch": 6.656, |
| "grad_norm": 2.046875, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 0.0836, |
| "mean_token_accuracy": 0.9799467187374831, |
| "num_tokens": 81390676.0, |
| "step": 4160 |
| }, |
| { |
| "entropy": 0.1643196914345026, |
| "epoch": 6.672, |
| "grad_norm": 2.125, |
| "learning_rate": 1.2957777777777779e-05, |
| "loss": 0.0869, |
| "mean_token_accuracy": 0.9793550491333007, |
| "num_tokens": 81584718.0, |
| "step": 4170 |
| }, |
| { |
| "entropy": 0.16305836467072368, |
| "epoch": 6.688, |
| "grad_norm": 2.125, |
| "learning_rate": 1.2935555555555556e-05, |
| "loss": 0.0882, |
| "mean_token_accuracy": 0.9788478314876556, |
| "num_tokens": 81774200.0, |
| "step": 4180 |
| }, |
| { |
| "entropy": 0.1591639060061425, |
| "epoch": 6.704, |
| "grad_norm": 2.265625, |
| "learning_rate": 1.2913333333333336e-05, |
| "loss": 0.084, |
| "mean_token_accuracy": 0.9798812437802553, |
| "num_tokens": 81974652.0, |
| "step": 4190 |
| }, |
| { |
| "entropy": 0.1584165041334927, |
| "epoch": 6.72, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.2891111111111112e-05, |
| "loss": 0.0847, |
| "mean_token_accuracy": 0.9802470825612545, |
| "num_tokens": 82173669.0, |
| "step": 4200 |
| }, |
| { |
| "epoch": 6.72, |
| "eval_biology_entropy": 0.4317656753063202, |
| "eval_biology_loss": 2.023526906967163, |
| "eval_biology_mean_token_accuracy": 0.6866904301643372, |
| "eval_biology_num_tokens": 82173669.0, |
| "eval_biology_runtime": 39.0742, |
| "eval_biology_samples_per_second": 12.796, |
| "eval_biology_steps_per_second": 3.199, |
| "step": 4200 |
| }, |
| { |
| "epoch": 6.72, |
| "eval_chemistry_entropy": 0.2700010150671005, |
| "eval_chemistry_loss": 1.274715542793274, |
| "eval_chemistry_mean_token_accuracy": 0.7817495450973511, |
| "eval_chemistry_num_tokens": 82173669.0, |
| "eval_chemistry_runtime": 48.6716, |
| "eval_chemistry_samples_per_second": 10.273, |
| "eval_chemistry_steps_per_second": 2.568, |
| "step": 4200 |
| }, |
| { |
| "entropy": 0.16326126656495035, |
| "epoch": 6.736, |
| "grad_norm": 2.140625, |
| "learning_rate": 1.2868888888888891e-05, |
| "loss": 0.0872, |
| "mean_token_accuracy": 0.9785537846386433, |
| "num_tokens": 82366613.0, |
| "step": 4210 |
| }, |
| { |
| "entropy": 0.16201097201555967, |
| "epoch": 6.752, |
| "grad_norm": 1.9609375, |
| "learning_rate": 1.2846666666666667e-05, |
| "loss": 0.0874, |
| "mean_token_accuracy": 0.9787759236991406, |
| "num_tokens": 82558119.0, |
| "step": 4220 |
| }, |
| { |
| "entropy": 0.1584986571688205, |
| "epoch": 6.768, |
| "grad_norm": 1.953125, |
| "learning_rate": 1.2824444444444446e-05, |
| "loss": 0.0838, |
| "mean_token_accuracy": 0.9800138369202613, |
| "num_tokens": 82758617.0, |
| "step": 4230 |
| }, |
| { |
| "entropy": 0.16598745151422917, |
| "epoch": 6.784, |
| "grad_norm": 1.9609375, |
| "learning_rate": 1.2802222222222222e-05, |
| "loss": 0.0855, |
| "mean_token_accuracy": 0.979158066958189, |
| "num_tokens": 82949626.0, |
| "step": 4240 |
| }, |
| { |
| "entropy": 0.15996336988173426, |
| "epoch": 6.8, |
| "grad_norm": 2.015625, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 0.086, |
| "mean_token_accuracy": 0.9797249909490346, |
| "num_tokens": 83144925.0, |
| "step": 4250 |
| }, |
| { |
| "entropy": 0.15937306629493833, |
| "epoch": 6.816, |
| "grad_norm": 2.15625, |
| "learning_rate": 1.2757777777777777e-05, |
| "loss": 0.0852, |
| "mean_token_accuracy": 0.9794587329030037, |
| "num_tokens": 83340639.0, |
| "step": 4260 |
| }, |
| { |
| "entropy": 0.16471727672033012, |
| "epoch": 6.832, |
| "grad_norm": 2.140625, |
| "learning_rate": 1.2735555555555557e-05, |
| "loss": 0.0899, |
| "mean_token_accuracy": 0.9784785751253366, |
| "num_tokens": 83528551.0, |
| "step": 4270 |
| }, |
| { |
| "entropy": 0.15601362506859004, |
| "epoch": 6.848, |
| "grad_norm": 2.28125, |
| "learning_rate": 1.2713333333333336e-05, |
| "loss": 0.0836, |
| "mean_token_accuracy": 0.9804382026195526, |
| "num_tokens": 83731837.0, |
| "step": 4280 |
| }, |
| { |
| "entropy": 0.16077870414592327, |
| "epoch": 6.864, |
| "grad_norm": 1.9375, |
| "learning_rate": 1.2691111111111112e-05, |
| "loss": 0.0847, |
| "mean_token_accuracy": 0.9797964643687009, |
| "num_tokens": 83930938.0, |
| "step": 4290 |
| }, |
| { |
| "entropy": 0.16511865155771374, |
| "epoch": 6.88, |
| "grad_norm": 2.703125, |
| "learning_rate": 1.2668888888888891e-05, |
| "loss": 0.0876, |
| "mean_token_accuracy": 0.9788475476205349, |
| "num_tokens": 84120848.0, |
| "step": 4300 |
| }, |
| { |
| "epoch": 6.88, |
| "eval_biology_entropy": 0.4311706922054291, |
| "eval_biology_loss": 2.0228519439697266, |
| "eval_biology_mean_token_accuracy": 0.6857039060592651, |
| "eval_biology_num_tokens": 84120848.0, |
| "eval_biology_runtime": 39.0592, |
| "eval_biology_samples_per_second": 12.801, |
| "eval_biology_steps_per_second": 3.2, |
| "step": 4300 |
| }, |
| { |
| "epoch": 6.88, |
| "eval_chemistry_entropy": 0.27077778267860414, |
| "eval_chemistry_loss": 1.2763019800186157, |
| "eval_chemistry_mean_token_accuracy": 0.7814345889091492, |
| "eval_chemistry_num_tokens": 84120848.0, |
| "eval_chemistry_runtime": 48.6615, |
| "eval_chemistry_samples_per_second": 10.275, |
| "eval_chemistry_steps_per_second": 2.569, |
| "step": 4300 |
| }, |
| { |
| "entropy": 0.16114730606786906, |
| "epoch": 6.896, |
| "grad_norm": 2.234375, |
| "learning_rate": 1.2646666666666667e-05, |
| "loss": 0.0849, |
| "mean_token_accuracy": 0.9798633921891451, |
| "num_tokens": 84315637.0, |
| "step": 4310 |
| }, |
| { |
| "entropy": 0.15914975721389055, |
| "epoch": 6.912, |
| "grad_norm": 2.140625, |
| "learning_rate": 1.2624444444444446e-05, |
| "loss": 0.0862, |
| "mean_token_accuracy": 0.9799407016485929, |
| "num_tokens": 84510176.0, |
| "step": 4320 |
| }, |
| { |
| "entropy": 0.1590470960829407, |
| "epoch": 6.928, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.2602222222222222e-05, |
| "loss": 0.0881, |
| "mean_token_accuracy": 0.978644073754549, |
| "num_tokens": 84709305.0, |
| "step": 4330 |
| }, |
| { |
| "entropy": 0.15750276539474725, |
| "epoch": 6.944, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 0.0834, |
| "mean_token_accuracy": 0.9799559228122234, |
| "num_tokens": 84912213.0, |
| "step": 4340 |
| }, |
| { |
| "entropy": 0.1642429924570024, |
| "epoch": 6.96, |
| "grad_norm": 2.171875, |
| "learning_rate": 1.2557777777777779e-05, |
| "loss": 0.0879, |
| "mean_token_accuracy": 0.9790333487093449, |
| "num_tokens": 85106675.0, |
| "step": 4350 |
| }, |
| { |
| "entropy": 0.16038998244330288, |
| "epoch": 6.976, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.2535555555555557e-05, |
| "loss": 0.0866, |
| "mean_token_accuracy": 0.9793150998651982, |
| "num_tokens": 85301501.0, |
| "step": 4360 |
| }, |
| { |
| "entropy": 0.1621239125728607, |
| "epoch": 6.992, |
| "grad_norm": 2.265625, |
| "learning_rate": 1.2513333333333334e-05, |
| "loss": 0.087, |
| "mean_token_accuracy": 0.9787631273269654, |
| "num_tokens": 85495488.0, |
| "step": 4370 |
| }, |
| { |
| "entropy": 0.14195599360391498, |
| "epoch": 7.008, |
| "grad_norm": 1.4765625, |
| "learning_rate": 1.2491111111111112e-05, |
| "loss": 0.0721, |
| "mean_token_accuracy": 0.9822378154844046, |
| "num_tokens": 85693979.0, |
| "step": 4380 |
| }, |
| { |
| "entropy": 0.11660091411322356, |
| "epoch": 7.024, |
| "grad_norm": 1.40625, |
| "learning_rate": 1.246888888888889e-05, |
| "loss": 0.0613, |
| "mean_token_accuracy": 0.9849191434681416, |
| "num_tokens": 85887341.0, |
| "step": 4390 |
| }, |
| { |
| "entropy": 0.11993917380459607, |
| "epoch": 7.04, |
| "grad_norm": 1.3671875, |
| "learning_rate": 1.2446666666666667e-05, |
| "loss": 0.0622, |
| "mean_token_accuracy": 0.9845501519739628, |
| "num_tokens": 86079618.0, |
| "step": 4400 |
| }, |
| { |
| "epoch": 7.04, |
| "eval_biology_entropy": 0.41136220598220824, |
| "eval_biology_loss": 2.076357841491699, |
| "eval_biology_mean_token_accuracy": 0.6870825448036194, |
| "eval_biology_num_tokens": 86079618.0, |
| "eval_biology_runtime": 39.0951, |
| "eval_biology_samples_per_second": 12.789, |
| "eval_biology_steps_per_second": 3.197, |
| "step": 4400 |
| }, |
| { |
| "epoch": 7.04, |
| "eval_chemistry_entropy": 0.2575183936357498, |
| "eval_chemistry_loss": 1.316386103630066, |
| "eval_chemistry_mean_token_accuracy": 0.7817704186439515, |
| "eval_chemistry_num_tokens": 86079618.0, |
| "eval_chemistry_runtime": 48.6834, |
| "eval_chemistry_samples_per_second": 10.27, |
| "eval_chemistry_steps_per_second": 2.568, |
| "step": 4400 |
| }, |
| { |
| "entropy": 0.1152312425430864, |
| "epoch": 7.056, |
| "grad_norm": 1.2578125, |
| "learning_rate": 1.2424444444444445e-05, |
| "loss": 0.0616, |
| "mean_token_accuracy": 0.985201733931899, |
| "num_tokens": 86278014.0, |
| "step": 4410 |
| }, |
| { |
| "entropy": 0.11724592591635882, |
| "epoch": 7.072, |
| "grad_norm": 1.390625, |
| "learning_rate": 1.2402222222222222e-05, |
| "loss": 0.0611, |
| "mean_token_accuracy": 0.9846691403537988, |
| "num_tokens": 86475886.0, |
| "step": 4420 |
| }, |
| { |
| "entropy": 0.11921753583010286, |
| "epoch": 7.088, |
| "grad_norm": 1.375, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 0.062, |
| "mean_token_accuracy": 0.9843826066702605, |
| "num_tokens": 86673786.0, |
| "step": 4430 |
| }, |
| { |
| "entropy": 0.1175499118398875, |
| "epoch": 7.104, |
| "grad_norm": 1.203125, |
| "learning_rate": 1.235777777777778e-05, |
| "loss": 0.0637, |
| "mean_token_accuracy": 0.9843353871256113, |
| "num_tokens": 86862992.0, |
| "step": 4440 |
| }, |
| { |
| "entropy": 0.11872103409841657, |
| "epoch": 7.12, |
| "grad_norm": 1.5546875, |
| "learning_rate": 1.2335555555555557e-05, |
| "loss": 0.0644, |
| "mean_token_accuracy": 0.9842873636633158, |
| "num_tokens": 87056495.0, |
| "step": 4450 |
| }, |
| { |
| "entropy": 0.11667176480405032, |
| "epoch": 7.136, |
| "grad_norm": 1.1953125, |
| "learning_rate": 1.2313333333333335e-05, |
| "loss": 0.0616, |
| "mean_token_accuracy": 0.9848171140998602, |
| "num_tokens": 87254100.0, |
| "step": 4460 |
| }, |
| { |
| "entropy": 0.11898361598141491, |
| "epoch": 7.152, |
| "grad_norm": 1.4375, |
| "learning_rate": 1.2291111111111112e-05, |
| "loss": 0.0629, |
| "mean_token_accuracy": 0.9844482876360416, |
| "num_tokens": 87448018.0, |
| "step": 4470 |
| }, |
| { |
| "entropy": 0.11841044360771776, |
| "epoch": 7.168, |
| "grad_norm": 1.59375, |
| "learning_rate": 1.226888888888889e-05, |
| "loss": 0.0623, |
| "mean_token_accuracy": 0.9847261656075716, |
| "num_tokens": 87648030.0, |
| "step": 4480 |
| }, |
| { |
| "entropy": 0.11289675291627646, |
| "epoch": 7.184, |
| "grad_norm": 1.5234375, |
| "learning_rate": 1.2246666666666667e-05, |
| "loss": 0.0607, |
| "mean_token_accuracy": 0.9850605838000774, |
| "num_tokens": 87849587.0, |
| "step": 4490 |
| }, |
| { |
| "entropy": 0.12062298045493662, |
| "epoch": 7.2, |
| "grad_norm": 1.25, |
| "learning_rate": 1.2224444444444445e-05, |
| "loss": 0.0641, |
| "mean_token_accuracy": 0.9840337805449962, |
| "num_tokens": 88041767.0, |
| "step": 4500 |
| }, |
| { |
| "epoch": 7.2, |
| "eval_biology_entropy": 0.40782594060897825, |
| "eval_biology_loss": 2.0909483432769775, |
| "eval_biology_mean_token_accuracy": 0.687857075214386, |
| "eval_biology_num_tokens": 88041767.0, |
| "eval_biology_runtime": 39.1341, |
| "eval_biology_samples_per_second": 12.777, |
| "eval_biology_steps_per_second": 3.194, |
| "step": 4500 |
| }, |
| { |
| "epoch": 7.2, |
| "eval_chemistry_entropy": 0.2562426084280014, |
| "eval_chemistry_loss": 1.3287914991378784, |
| "eval_chemistry_mean_token_accuracy": 0.7817574729919433, |
| "eval_chemistry_num_tokens": 88041767.0, |
| "eval_chemistry_runtime": 48.6523, |
| "eval_chemistry_samples_per_second": 10.277, |
| "eval_chemistry_steps_per_second": 2.569, |
| "step": 4500 |
| }, |
| { |
| "entropy": 0.12015473688952624, |
| "epoch": 7.216, |
| "grad_norm": 1.609375, |
| "learning_rate": 1.2202222222222224e-05, |
| "loss": 0.0641, |
| "mean_token_accuracy": 0.9842524129897356, |
| "num_tokens": 88237413.0, |
| "step": 4510 |
| }, |
| { |
| "entropy": 0.11671558041125536, |
| "epoch": 7.232, |
| "grad_norm": 1.71875, |
| "learning_rate": 1.218e-05, |
| "loss": 0.0626, |
| "mean_token_accuracy": 0.9845516506582499, |
| "num_tokens": 88435523.0, |
| "step": 4520 |
| }, |
| { |
| "entropy": 0.12031128369271755, |
| "epoch": 7.248, |
| "grad_norm": 1.4375, |
| "learning_rate": 1.215777777777778e-05, |
| "loss": 0.0646, |
| "mean_token_accuracy": 0.9843074388802051, |
| "num_tokens": 88627287.0, |
| "step": 4530 |
| }, |
| { |
| "entropy": 0.12133477986790239, |
| "epoch": 7.264, |
| "grad_norm": 1.2578125, |
| "learning_rate": 1.2135555555555556e-05, |
| "loss": 0.0648, |
| "mean_token_accuracy": 0.9840338047593832, |
| "num_tokens": 88816114.0, |
| "step": 4540 |
| }, |
| { |
| "entropy": 0.11867788145318628, |
| "epoch": 7.28, |
| "grad_norm": 1.4453125, |
| "learning_rate": 1.2113333333333335e-05, |
| "loss": 0.0635, |
| "mean_token_accuracy": 0.9847275119274854, |
| "num_tokens": 89014170.0, |
| "step": 4550 |
| }, |
| { |
| "entropy": 0.11938378578051925, |
| "epoch": 7.296, |
| "grad_norm": 1.6484375, |
| "learning_rate": 1.209111111111111e-05, |
| "loss": 0.0645, |
| "mean_token_accuracy": 0.9842457465827466, |
| "num_tokens": 89208392.0, |
| "step": 4560 |
| }, |
| { |
| "entropy": 0.11515432372689247, |
| "epoch": 7.312, |
| "grad_norm": 1.09375, |
| "learning_rate": 1.206888888888889e-05, |
| "loss": 0.0628, |
| "mean_token_accuracy": 0.9846439093351365, |
| "num_tokens": 89410053.0, |
| "step": 4570 |
| }, |
| { |
| "entropy": 0.12289352240040899, |
| "epoch": 7.328, |
| "grad_norm": 1.4609375, |
| "learning_rate": 1.204666666666667e-05, |
| "loss": 0.066, |
| "mean_token_accuracy": 0.983601363748312, |
| "num_tokens": 89599409.0, |
| "step": 4580 |
| }, |
| { |
| "entropy": 0.12188131813891231, |
| "epoch": 7.344, |
| "grad_norm": 1.4140625, |
| "learning_rate": 1.2024444444444445e-05, |
| "loss": 0.0658, |
| "mean_token_accuracy": 0.9839568357914686, |
| "num_tokens": 89790090.0, |
| "step": 4590 |
| }, |
| { |
| "entropy": 0.11989519495982677, |
| "epoch": 7.36, |
| "grad_norm": 1.203125, |
| "learning_rate": 1.2002222222222225e-05, |
| "loss": 0.0653, |
| "mean_token_accuracy": 0.9840534634888172, |
| "num_tokens": 89980264.0, |
| "step": 4600 |
| }, |
| { |
| "epoch": 7.36, |
| "eval_biology_entropy": 0.40814528465270994, |
| "eval_biology_loss": 2.095086097717285, |
| "eval_biology_mean_token_accuracy": 0.6874291839599609, |
| "eval_biology_num_tokens": 89980264.0, |
| "eval_biology_runtime": 39.4682, |
| "eval_biology_samples_per_second": 12.668, |
| "eval_biology_steps_per_second": 3.167, |
| "step": 4600 |
| }, |
| { |
| "epoch": 7.36, |
| "eval_chemistry_entropy": 0.25585675418376924, |
| "eval_chemistry_loss": 1.3295270204544067, |
| "eval_chemistry_mean_token_accuracy": 0.7816342005729675, |
| "eval_chemistry_num_tokens": 89980264.0, |
| "eval_chemistry_runtime": 49.1953, |
| "eval_chemistry_samples_per_second": 10.164, |
| "eval_chemistry_steps_per_second": 2.541, |
| "step": 4600 |
| }, |
| { |
| "entropy": 0.11714847502298653, |
| "epoch": 7.376, |
| "grad_norm": 1.4140625, |
| "learning_rate": 1.198e-05, |
| "loss": 0.0634, |
| "mean_token_accuracy": 0.9846220836043358, |
| "num_tokens": 90182815.0, |
| "step": 4610 |
| }, |
| { |
| "entropy": 0.11851101261563599, |
| "epoch": 7.392, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.195777777777778e-05, |
| "loss": 0.0661, |
| "mean_token_accuracy": 0.9842429973185063, |
| "num_tokens": 90381414.0, |
| "step": 4620 |
| }, |
| { |
| "entropy": 0.11742206565104425, |
| "epoch": 7.408, |
| "grad_norm": 1.3671875, |
| "learning_rate": 1.1935555555555556e-05, |
| "loss": 0.0643, |
| "mean_token_accuracy": 0.9843564033508301, |
| "num_tokens": 90576675.0, |
| "step": 4630 |
| }, |
| { |
| "entropy": 0.11906451536342502, |
| "epoch": 7.424, |
| "grad_norm": 1.34375, |
| "learning_rate": 1.1913333333333335e-05, |
| "loss": 0.0639, |
| "mean_token_accuracy": 0.9840137660503387, |
| "num_tokens": 90772883.0, |
| "step": 4640 |
| }, |
| { |
| "entropy": 0.11538688251748681, |
| "epoch": 7.44, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.1891111111111111e-05, |
| "loss": 0.0615, |
| "mean_token_accuracy": 0.9849125079810619, |
| "num_tokens": 90979968.0, |
| "step": 4650 |
| }, |
| { |
| "entropy": 0.1188006397569552, |
| "epoch": 7.456, |
| "grad_norm": 1.28125, |
| "learning_rate": 1.186888888888889e-05, |
| "loss": 0.0652, |
| "mean_token_accuracy": 0.9839730773121118, |
| "num_tokens": 91173673.0, |
| "step": 4660 |
| }, |
| { |
| "entropy": 0.1222672678064555, |
| "epoch": 7.4719999999999995, |
| "grad_norm": 1.3984375, |
| "learning_rate": 1.1846666666666668e-05, |
| "loss": 0.0653, |
| "mean_token_accuracy": 0.9837797362357378, |
| "num_tokens": 91367657.0, |
| "step": 4670 |
| }, |
| { |
| "entropy": 0.12343765725381672, |
| "epoch": 7.4879999999999995, |
| "grad_norm": 1.546875, |
| "learning_rate": 1.1824444444444445e-05, |
| "loss": 0.0673, |
| "mean_token_accuracy": 0.9835902791470289, |
| "num_tokens": 91556884.0, |
| "step": 4680 |
| }, |
| { |
| "entropy": 0.1220437603071332, |
| "epoch": 7.504, |
| "grad_norm": 1.5234375, |
| "learning_rate": 1.1802222222222223e-05, |
| "loss": 0.0657, |
| "mean_token_accuracy": 0.983706547319889, |
| "num_tokens": 91746927.0, |
| "step": 4690 |
| }, |
| { |
| "entropy": 0.1192731964867562, |
| "epoch": 7.52, |
| "grad_norm": 1.3203125, |
| "learning_rate": 1.178e-05, |
| "loss": 0.0641, |
| "mean_token_accuracy": 0.9843227826058865, |
| "num_tokens": 91945054.0, |
| "step": 4700 |
| }, |
| { |
| "epoch": 7.52, |
| "eval_biology_entropy": 0.4071502583026886, |
| "eval_biology_loss": 2.104748010635376, |
| "eval_biology_mean_token_accuracy": 0.6872874584197998, |
| "eval_biology_num_tokens": 91945054.0, |
| "eval_biology_runtime": 39.5007, |
| "eval_biology_samples_per_second": 12.658, |
| "eval_biology_steps_per_second": 3.165, |
| "step": 4700 |
| }, |
| { |
| "epoch": 7.52, |
| "eval_chemistry_entropy": 0.2555647749900818, |
| "eval_chemistry_loss": 1.3347607851028442, |
| "eval_chemistry_mean_token_accuracy": 0.7812189860343933, |
| "eval_chemistry_num_tokens": 91945054.0, |
| "eval_chemistry_runtime": 49.1833, |
| "eval_chemistry_samples_per_second": 10.166, |
| "eval_chemistry_steps_per_second": 2.542, |
| "step": 4700 |
| }, |
| { |
| "entropy": 0.12297428376041353, |
| "epoch": 7.536, |
| "grad_norm": 1.6171875, |
| "learning_rate": 1.1757777777777778e-05, |
| "loss": 0.068, |
| "mean_token_accuracy": 0.9833006031811238, |
| "num_tokens": 92131859.0, |
| "step": 4710 |
| }, |
| { |
| "entropy": 0.1182135074865073, |
| "epoch": 7.552, |
| "grad_norm": 1.3515625, |
| "learning_rate": 1.1735555555555556e-05, |
| "loss": 0.0644, |
| "mean_token_accuracy": 0.9843916691839695, |
| "num_tokens": 92330048.0, |
| "step": 4720 |
| }, |
| { |
| "entropy": 0.12027381157968194, |
| "epoch": 7.568, |
| "grad_norm": 1.703125, |
| "learning_rate": 1.1713333333333334e-05, |
| "loss": 0.0662, |
| "mean_token_accuracy": 0.9834937732666731, |
| "num_tokens": 92523606.0, |
| "step": 4730 |
| }, |
| { |
| "entropy": 0.11872511534020305, |
| "epoch": 7.584, |
| "grad_norm": 1.3125, |
| "learning_rate": 1.1691111111111113e-05, |
| "loss": 0.064, |
| "mean_token_accuracy": 0.9840778723359108, |
| "num_tokens": 92722289.0, |
| "step": 4740 |
| }, |
| { |
| "entropy": 0.11951107773929834, |
| "epoch": 7.6, |
| "grad_norm": 1.3359375, |
| "learning_rate": 1.1668888888888889e-05, |
| "loss": 0.0642, |
| "mean_token_accuracy": 0.9840859781950713, |
| "num_tokens": 92919710.0, |
| "step": 4750 |
| }, |
| { |
| "entropy": 0.11732456265017391, |
| "epoch": 7.616, |
| "grad_norm": 1.5078125, |
| "learning_rate": 1.1646666666666668e-05, |
| "loss": 0.0639, |
| "mean_token_accuracy": 0.9842873651534318, |
| "num_tokens": 93117203.0, |
| "step": 4760 |
| }, |
| { |
| "entropy": 0.12109015956521034, |
| "epoch": 7.632, |
| "grad_norm": 1.9609375, |
| "learning_rate": 1.1624444444444446e-05, |
| "loss": 0.0657, |
| "mean_token_accuracy": 0.9837942615151405, |
| "num_tokens": 93315418.0, |
| "step": 4770 |
| }, |
| { |
| "entropy": 0.12157906177453696, |
| "epoch": 7.648, |
| "grad_norm": 1.4765625, |
| "learning_rate": 1.1602222222222223e-05, |
| "loss": 0.0652, |
| "mean_token_accuracy": 0.9839387238025665, |
| "num_tokens": 93508661.0, |
| "step": 4780 |
| }, |
| { |
| "entropy": 0.12103188117034733, |
| "epoch": 7.664, |
| "grad_norm": 1.2109375, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 0.0666, |
| "mean_token_accuracy": 0.9840183984488249, |
| "num_tokens": 93698986.0, |
| "step": 4790 |
| }, |
| { |
| "entropy": 0.11844088467769325, |
| "epoch": 7.68, |
| "grad_norm": 1.40625, |
| "learning_rate": 1.1557777777777779e-05, |
| "loss": 0.0634, |
| "mean_token_accuracy": 0.9844320382922888, |
| "num_tokens": 93898443.0, |
| "step": 4800 |
| }, |
| { |
| "epoch": 7.68, |
| "eval_biology_entropy": 0.4091689692735672, |
| "eval_biology_loss": 2.101299285888672, |
| "eval_biology_mean_token_accuracy": 0.6870952935218811, |
| "eval_biology_num_tokens": 93898443.0, |
| "eval_biology_runtime": 39.5716, |
| "eval_biology_samples_per_second": 12.635, |
| "eval_biology_steps_per_second": 3.159, |
| "step": 4800 |
| }, |
| { |
| "epoch": 7.68, |
| "eval_chemistry_entropy": 0.25760244596004483, |
| "eval_chemistry_loss": 1.3335673809051514, |
| "eval_chemistry_mean_token_accuracy": 0.7808964705467224, |
| "eval_chemistry_num_tokens": 93898443.0, |
| "eval_chemistry_runtime": 49.0921, |
| "eval_chemistry_samples_per_second": 10.185, |
| "eval_chemistry_steps_per_second": 2.546, |
| "step": 4800 |
| }, |
| { |
| "entropy": 0.1192777092102915, |
| "epoch": 7.696, |
| "grad_norm": 1.6015625, |
| "learning_rate": 1.1535555555555556e-05, |
| "loss": 0.0642, |
| "mean_token_accuracy": 0.9843660224229097, |
| "num_tokens": 94099003.0, |
| "step": 4810 |
| }, |
| { |
| "entropy": 0.12081068973056971, |
| "epoch": 7.712, |
| "grad_norm": 1.84375, |
| "learning_rate": 1.1513333333333334e-05, |
| "loss": 0.0666, |
| "mean_token_accuracy": 0.983449924737215, |
| "num_tokens": 94290532.0, |
| "step": 4820 |
| }, |
| { |
| "entropy": 0.11797172902151942, |
| "epoch": 7.728, |
| "grad_norm": 1.359375, |
| "learning_rate": 1.1491111111111113e-05, |
| "loss": 0.0638, |
| "mean_token_accuracy": 0.9843352895230055, |
| "num_tokens": 94491294.0, |
| "step": 4830 |
| }, |
| { |
| "entropy": 0.11733483774587512, |
| "epoch": 7.744, |
| "grad_norm": 1.2734375, |
| "learning_rate": 1.1468888888888889e-05, |
| "loss": 0.0637, |
| "mean_token_accuracy": 0.9844149738550186, |
| "num_tokens": 94690330.0, |
| "step": 4840 |
| }, |
| { |
| "entropy": 0.11756395041011274, |
| "epoch": 7.76, |
| "grad_norm": 1.5234375, |
| "learning_rate": 1.1446666666666668e-05, |
| "loss": 0.0656, |
| "mean_token_accuracy": 0.9841559953987599, |
| "num_tokens": 94888279.0, |
| "step": 4850 |
| }, |
| { |
| "entropy": 0.12167371204122901, |
| "epoch": 7.776, |
| "grad_norm": 1.265625, |
| "learning_rate": 1.1424444444444444e-05, |
| "loss": 0.0651, |
| "mean_token_accuracy": 0.9838530842214823, |
| "num_tokens": 95084013.0, |
| "step": 4860 |
| }, |
| { |
| "entropy": 0.12155953082256019, |
| "epoch": 7.792, |
| "grad_norm": 1.4375, |
| "learning_rate": 1.1402222222222224e-05, |
| "loss": 0.066, |
| "mean_token_accuracy": 0.9837955456227064, |
| "num_tokens": 95276019.0, |
| "step": 4870 |
| }, |
| { |
| "entropy": 0.12080176440067589, |
| "epoch": 7.808, |
| "grad_norm": 1.703125, |
| "learning_rate": 1.138e-05, |
| "loss": 0.0659, |
| "mean_token_accuracy": 0.9838258396834135, |
| "num_tokens": 95470187.0, |
| "step": 4880 |
| }, |
| { |
| "entropy": 0.12329842476174235, |
| "epoch": 7.824, |
| "grad_norm": 1.375, |
| "learning_rate": 1.1357777777777779e-05, |
| "loss": 0.0677, |
| "mean_token_accuracy": 0.9834286782890558, |
| "num_tokens": 95659931.0, |
| "step": 4890 |
| }, |
| { |
| "entropy": 0.11845089322887362, |
| "epoch": 7.84, |
| "grad_norm": 1.25, |
| "learning_rate": 1.1335555555555558e-05, |
| "loss": 0.065, |
| "mean_token_accuracy": 0.9840061299502849, |
| "num_tokens": 95856835.0, |
| "step": 4900 |
| }, |
| { |
| "epoch": 7.84, |
| "eval_biology_entropy": 0.40712755966186526, |
| "eval_biology_loss": 2.1040472984313965, |
| "eval_biology_mean_token_accuracy": 0.6877834839820862, |
| "eval_biology_num_tokens": 95856835.0, |
| "eval_biology_runtime": 39.1004, |
| "eval_biology_samples_per_second": 12.788, |
| "eval_biology_steps_per_second": 3.197, |
| "step": 4900 |
| }, |
| { |
| "epoch": 7.84, |
| "eval_chemistry_entropy": 0.2551681501865387, |
| "eval_chemistry_loss": 1.3381378650665283, |
| "eval_chemistry_mean_token_accuracy": 0.78116206407547, |
| "eval_chemistry_num_tokens": 95856835.0, |
| "eval_chemistry_runtime": 48.7019, |
| "eval_chemistry_samples_per_second": 10.267, |
| "eval_chemistry_steps_per_second": 2.567, |
| "step": 4900 |
| }, |
| { |
| "entropy": 0.11883879867382348, |
| "epoch": 7.856, |
| "grad_norm": 1.546875, |
| "learning_rate": 1.1313333333333334e-05, |
| "loss": 0.0651, |
| "mean_token_accuracy": 0.983873812481761, |
| "num_tokens": 96051765.0, |
| "step": 4910 |
| }, |
| { |
| "entropy": 0.12126617752946914, |
| "epoch": 7.872, |
| "grad_norm": 1.4609375, |
| "learning_rate": 1.1291111111111113e-05, |
| "loss": 0.0647, |
| "mean_token_accuracy": 0.9838901285082102, |
| "num_tokens": 96251297.0, |
| "step": 4920 |
| }, |
| { |
| "entropy": 0.1180964884813875, |
| "epoch": 7.888, |
| "grad_norm": 1.0390625, |
| "learning_rate": 1.126888888888889e-05, |
| "loss": 0.0632, |
| "mean_token_accuracy": 0.9843307383358478, |
| "num_tokens": 96452255.0, |
| "step": 4930 |
| }, |
| { |
| "entropy": 0.12124264547601342, |
| "epoch": 7.904, |
| "grad_norm": 1.1875, |
| "learning_rate": 1.1246666666666669e-05, |
| "loss": 0.0655, |
| "mean_token_accuracy": 0.983839163556695, |
| "num_tokens": 96643810.0, |
| "step": 4940 |
| }, |
| { |
| "entropy": 0.12158664194867015, |
| "epoch": 7.92, |
| "grad_norm": 1.5859375, |
| "learning_rate": 1.1224444444444444e-05, |
| "loss": 0.0656, |
| "mean_token_accuracy": 0.9837770704180002, |
| "num_tokens": 96838559.0, |
| "step": 4950 |
| }, |
| { |
| "entropy": 0.12063693962991237, |
| "epoch": 7.936, |
| "grad_norm": 1.625, |
| "learning_rate": 1.1202222222222224e-05, |
| "loss": 0.0675, |
| "mean_token_accuracy": 0.983657094463706, |
| "num_tokens": 97027366.0, |
| "step": 4960 |
| }, |
| { |
| "entropy": 0.12248484031297266, |
| "epoch": 7.952, |
| "grad_norm": 1.4140625, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 0.0653, |
| "mean_token_accuracy": 0.9837716151028871, |
| "num_tokens": 97222139.0, |
| "step": 4970 |
| }, |
| { |
| "entropy": 0.12115419898182153, |
| "epoch": 7.968, |
| "grad_norm": 1.359375, |
| "learning_rate": 1.1157777777777779e-05, |
| "loss": 0.0672, |
| "mean_token_accuracy": 0.9834024798125028, |
| "num_tokens": 97413100.0, |
| "step": 4980 |
| }, |
| { |
| "entropy": 0.11634420100599527, |
| "epoch": 7.984, |
| "grad_norm": 1.5078125, |
| "learning_rate": 1.1135555555555557e-05, |
| "loss": 0.0631, |
| "mean_token_accuracy": 0.9845709603279829, |
| "num_tokens": 97618719.0, |
| "step": 4990 |
| }, |
| { |
| "entropy": 0.11867476101033389, |
| "epoch": 8.0, |
| "grad_norm": 1.5390625, |
| "learning_rate": 1.1113333333333334e-05, |
| "loss": 0.0638, |
| "mean_token_accuracy": 0.9840520035475493, |
| "num_tokens": 97822176.0, |
| "step": 5000 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_biology_entropy": 0.4059671058654785, |
| "eval_biology_loss": 2.107717752456665, |
| "eval_biology_mean_token_accuracy": 0.6878666534423828, |
| "eval_biology_num_tokens": 97822176.0, |
| "eval_biology_runtime": 39.0889, |
| "eval_biology_samples_per_second": 12.791, |
| "eval_biology_steps_per_second": 3.198, |
| "step": 5000 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_chemistry_entropy": 0.25426543402671814, |
| "eval_chemistry_loss": 1.3373106718063354, |
| "eval_chemistry_mean_token_accuracy": 0.7816811356544494, |
| "eval_chemistry_num_tokens": 97822176.0, |
| "eval_chemistry_runtime": 48.507, |
| "eval_chemistry_samples_per_second": 10.308, |
| "eval_chemistry_steps_per_second": 2.577, |
| "step": 5000 |
| } |
| ], |
| "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": 8.215055580334719e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
