| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.016407563886951883, | |
| "eval_steps": 50, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.203781943475943e-05, | |
| "eval_loss": 2.369717836380005, | |
| "eval_runtime": 120.3697, | |
| "eval_samples_per_second": 42.644, | |
| "eval_steps_per_second": 21.326, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0004101890971737971, | |
| "grad_norm": 2.2281951904296875, | |
| "learning_rate": 5e-05, | |
| "loss": 2.1519, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0008203781943475942, | |
| "grad_norm": 1.7547389268875122, | |
| "learning_rate": 0.0001, | |
| "loss": 2.0718, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0012305672915213913, | |
| "grad_norm": 1.2035329341888428, | |
| "learning_rate": 9.98292246503335e-05, | |
| "loss": 1.6792, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0016407563886951884, | |
| "grad_norm": 1.3260753154754639, | |
| "learning_rate": 9.931806517013612e-05, | |
| "loss": 1.5188, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0020509454858689854, | |
| "grad_norm": 1.4489283561706543, | |
| "learning_rate": 9.847001329696653e-05, | |
| "loss": 1.3314, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0024611345830427827, | |
| "grad_norm": 1.5003387928009033, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 1.3347, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00287132368021658, | |
| "grad_norm": 1.1698554754257202, | |
| "learning_rate": 9.578866633275288e-05, | |
| "loss": 1.0366, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0032815127773903768, | |
| "grad_norm": 1.1663718223571777, | |
| "learning_rate": 9.397368756032445e-05, | |
| "loss": 1.0537, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.003691701874564174, | |
| "grad_norm": 0.9623744487762451, | |
| "learning_rate": 9.185832391312644e-05, | |
| "loss": 1.0906, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004101890971737971, | |
| "grad_norm": 1.3753273487091064, | |
| "learning_rate": 8.945702546981969e-05, | |
| "loss": 0.9883, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.004101890971737971, | |
| "eval_loss": 1.1852165460586548, | |
| "eval_runtime": 120.0085, | |
| "eval_samples_per_second": 42.772, | |
| "eval_steps_per_second": 21.39, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0045120800689117686, | |
| "grad_norm": 1.2335317134857178, | |
| "learning_rate": 8.678619553365659e-05, | |
| "loss": 1.2677, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.004922269166085565, | |
| "grad_norm": 0.86533522605896, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 1.1786, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.005332458263259362, | |
| "grad_norm": 1.1198489665985107, | |
| "learning_rate": 8.07106356344834e-05, | |
| "loss": 1.1856, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.00574264736043316, | |
| "grad_norm": 1.0070888996124268, | |
| "learning_rate": 7.734740790612136e-05, | |
| "loss": 1.1923, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.006152836457606957, | |
| "grad_norm": 0.9945841431617737, | |
| "learning_rate": 7.379736965185368e-05, | |
| "loss": 1.0521, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0065630255547807536, | |
| "grad_norm": 1.007394552230835, | |
| "learning_rate": 7.008477123264848e-05, | |
| "loss": 1.2133, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.006973214651954551, | |
| "grad_norm": 1.1531294584274292, | |
| "learning_rate": 6.623497346023418e-05, | |
| "loss": 1.1182, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.007383403749128348, | |
| "grad_norm": 1.2772388458251953, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 1.2614, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.007793592846302145, | |
| "grad_norm": 1.2343947887420654, | |
| "learning_rate": 5.8229729514036705e-05, | |
| "loss": 1.1619, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.008203781943475942, | |
| "grad_norm": 1.0785244703292847, | |
| "learning_rate": 5.4128967273616625e-05, | |
| "loss": 0.9675, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.008203781943475942, | |
| "eval_loss": 1.1615369319915771, | |
| "eval_runtime": 120.0295, | |
| "eval_samples_per_second": 42.764, | |
| "eval_steps_per_second": 21.386, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.00861397104064974, | |
| "grad_norm": 1.1045833826065063, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2067, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.009024160137823537, | |
| "grad_norm": 1.2426618337631226, | |
| "learning_rate": 4.5871032726383386e-05, | |
| "loss": 1.1525, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.009434349234997334, | |
| "grad_norm": 1.1180200576782227, | |
| "learning_rate": 4.17702704859633e-05, | |
| "loss": 1.167, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.00984453833217113, | |
| "grad_norm": 1.020121455192566, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 1.0811, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.010254727429344928, | |
| "grad_norm": 0.9257335066795349, | |
| "learning_rate": 3.3765026539765834e-05, | |
| "loss": 1.1457, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.010664916526518724, | |
| "grad_norm": 1.2891005277633667, | |
| "learning_rate": 2.991522876735154e-05, | |
| "loss": 1.1386, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.011075105623692523, | |
| "grad_norm": 0.9600415229797363, | |
| "learning_rate": 2.6202630348146324e-05, | |
| "loss": 1.095, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.01148529472086632, | |
| "grad_norm": 0.9995129704475403, | |
| "learning_rate": 2.2652592093878666e-05, | |
| "loss": 1.0003, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.011895483818040117, | |
| "grad_norm": 1.0976762771606445, | |
| "learning_rate": 1.928936436551661e-05, | |
| "loss": 1.0082, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.012305672915213913, | |
| "grad_norm": 1.282447338104248, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 1.1959, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.012305672915213913, | |
| "eval_loss": 1.1480756998062134, | |
| "eval_runtime": 120.1971, | |
| "eval_samples_per_second": 42.705, | |
| "eval_steps_per_second": 21.357, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.01271586201238771, | |
| "grad_norm": 0.9905368089675903, | |
| "learning_rate": 1.3213804466343421e-05, | |
| "loss": 1.2891, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.013126051109561507, | |
| "grad_norm": 1.3122817277908325, | |
| "learning_rate": 1.0542974530180327e-05, | |
| "loss": 1.2428, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.013536240206735306, | |
| "grad_norm": 1.2511900663375854, | |
| "learning_rate": 8.141676086873572e-06, | |
| "loss": 1.1483, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.013946429303909103, | |
| "grad_norm": 1.1989154815673828, | |
| "learning_rate": 6.026312439675552e-06, | |
| "loss": 1.2854, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.0143566184010829, | |
| "grad_norm": 1.10850989818573, | |
| "learning_rate": 4.2113336672471245e-06, | |
| "loss": 1.2867, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.014766807498256696, | |
| "grad_norm": 1.170858383178711, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 1.3531, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.015176996595430493, | |
| "grad_norm": 1.1136072874069214, | |
| "learning_rate": 1.5299867030334814e-06, | |
| "loss": 1.1354, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.01558718569260429, | |
| "grad_norm": 1.0280817747116089, | |
| "learning_rate": 6.819348298638839e-07, | |
| "loss": 1.0995, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.01599737478977809, | |
| "grad_norm": 1.445446252822876, | |
| "learning_rate": 1.7077534966650766e-07, | |
| "loss": 1.0685, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.016407563886951883, | |
| "grad_norm": 1.2722431421279907, | |
| "learning_rate": 0.0, | |
| "loss": 1.2453, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.016407563886951883, | |
| "eval_loss": 1.1455199718475342, | |
| "eval_runtime": 120.1974, | |
| "eval_samples_per_second": 42.705, | |
| "eval_steps_per_second": 21.357, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.02316791300096e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |