| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.09082652134423251, | |
| "eval_steps": 9, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0018165304268846503, | |
| "grad_norm": 0.3600485622882843, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4664, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0018165304268846503, | |
| "eval_loss": 0.4619494378566742, | |
| "eval_runtime": 108.3327, | |
| "eval_samples_per_second": 4.283, | |
| "eval_steps_per_second": 0.535, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0036330608537693005, | |
| "grad_norm": 0.33697664737701416, | |
| "learning_rate": 2e-05, | |
| "loss": 0.4766, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.005449591280653951, | |
| "grad_norm": 0.3403802812099457, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4548, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.007266121707538601, | |
| "grad_norm": 0.3392443060874939, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4661, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.009082652134423252, | |
| "grad_norm": 0.39870426058769226, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4286, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.010899182561307902, | |
| "grad_norm": 0.3802020847797394, | |
| "learning_rate": 6e-05, | |
| "loss": 0.4409, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.012715712988192553, | |
| "grad_norm": 0.3801540732383728, | |
| "learning_rate": 7e-05, | |
| "loss": 0.4209, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.014532243415077202, | |
| "grad_norm": 0.38932061195373535, | |
| "learning_rate": 8e-05, | |
| "loss": 0.4188, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01634877384196185, | |
| "grad_norm": 0.3350403308868408, | |
| "learning_rate": 9e-05, | |
| "loss": 0.396, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01634877384196185, | |
| "eval_loss": 0.3949369788169861, | |
| "eval_runtime": 108.084, | |
| "eval_samples_per_second": 4.293, | |
| "eval_steps_per_second": 0.537, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.018165304268846504, | |
| "grad_norm": 0.33399584889411926, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3662, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.019981834695731154, | |
| "grad_norm": 0.3276408314704895, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.3859, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.021798365122615803, | |
| "grad_norm": 0.3103690445423126, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.3512, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.023614895549500452, | |
| "grad_norm": 0.45643332600593567, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.3913, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.025431425976385105, | |
| "grad_norm": 0.38945508003234863, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.3104, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.027247956403269755, | |
| "grad_norm": 0.33456993103027344, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.3705, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.029064486830154404, | |
| "grad_norm": 0.4321489632129669, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.3217, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.030881017257039057, | |
| "grad_norm": 0.3709757924079895, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.3113, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0326975476839237, | |
| "grad_norm": 0.33215710520744324, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.3293, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0326975476839237, | |
| "eval_loss": 0.32641759514808655, | |
| "eval_runtime": 108.5742, | |
| "eval_samples_per_second": 4.274, | |
| "eval_steps_per_second": 0.534, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03451407811080836, | |
| "grad_norm": 0.40216928720474243, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.385, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.03633060853769301, | |
| "grad_norm": 0.3322986662387848, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.2936, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03814713896457766, | |
| "grad_norm": 0.339769572019577, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.329, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03996366939146231, | |
| "grad_norm": 0.34345075488090515, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.2768, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.04178019981834696, | |
| "grad_norm": 0.3251555860042572, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.2831, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.043596730245231606, | |
| "grad_norm": 0.35843387246131897, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.402, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.045413260672116255, | |
| "grad_norm": 0.31343409419059753, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.2751, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.047229791099000905, | |
| "grad_norm": 0.3558681607246399, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.3265, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.04904632152588556, | |
| "grad_norm": 0.30473554134368896, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.241, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.04904632152588556, | |
| "eval_loss": 0.30456194281578064, | |
| "eval_runtime": 108.8274, | |
| "eval_samples_per_second": 4.264, | |
| "eval_steps_per_second": 0.533, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.05086285195277021, | |
| "grad_norm": 0.4298734962940216, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.2689, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.05267938237965486, | |
| "grad_norm": 0.3488064110279083, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 0.2775, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.05449591280653951, | |
| "grad_norm": 0.4139954745769501, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.3113, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05631244323342416, | |
| "grad_norm": 0.3354235887527466, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.3051, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.05812897366030881, | |
| "grad_norm": 0.41172316670417786, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.3233, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.05994550408719346, | |
| "grad_norm": 0.3322085738182068, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.3209, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.061762034514078114, | |
| "grad_norm": 0.3583935797214508, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.3255, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.06357856494096276, | |
| "grad_norm": 0.2800469696521759, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.3104, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0653950953678474, | |
| "grad_norm": 0.4980929493904114, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.3561, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0653950953678474, | |
| "eval_loss": 0.2919258773326874, | |
| "eval_runtime": 108.6083, | |
| "eval_samples_per_second": 4.272, | |
| "eval_steps_per_second": 0.534, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.06721162579473206, | |
| "grad_norm": 0.3218163549900055, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.2759, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.06902815622161672, | |
| "grad_norm": 0.29215186834335327, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.3034, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.07084468664850137, | |
| "grad_norm": 0.3236890733242035, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.278, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.07266121707538602, | |
| "grad_norm": 0.3118452727794647, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.3184, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07447774750227067, | |
| "grad_norm": 0.2818981111049652, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.2823, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.07629427792915532, | |
| "grad_norm": 0.2725626230239868, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.2795, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.07811080835603997, | |
| "grad_norm": 0.28230541944503784, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.2023, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.07992733878292461, | |
| "grad_norm": 0.24861739575862885, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.2655, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.08174386920980926, | |
| "grad_norm": 0.29663726687431335, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.2904, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.08174386920980926, | |
| "eval_loss": 0.2852410078048706, | |
| "eval_runtime": 108.7895, | |
| "eval_samples_per_second": 4.265, | |
| "eval_steps_per_second": 0.533, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.08356039963669391, | |
| "grad_norm": 0.27381181716918945, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.2109, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.08537693006357856, | |
| "grad_norm": 0.3276028335094452, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.3141, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.08719346049046321, | |
| "grad_norm": 0.2625039219856262, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.2521, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.08900999091734786, | |
| "grad_norm": 0.31556904315948486, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.2891, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.09082652134423251, | |
| "grad_norm": 0.2701485753059387, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.2474, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.19631055627223e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |