| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 171, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.08771929824561403, |
| "grad_norm": 1.3670965433120728, |
| "learning_rate": 1.3333333333333333e-05, |
| "loss": 0.0503, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.17543859649122806, |
| "grad_norm": 1.5942356586456299, |
| "learning_rate": 1.9983691039261358e-05, |
| "loss": 0.0734, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.2631578947368421, |
| "grad_norm": 1.8806111812591553, |
| "learning_rate": 1.9884217748453625e-05, |
| "loss": 0.0809, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.3508771929824561, |
| "grad_norm": 2.370424270629883, |
| "learning_rate": 1.9695231451106914e-05, |
| "loss": 0.1011, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.43859649122807015, |
| "grad_norm": 1.674837589263916, |
| "learning_rate": 1.941844363639525e-05, |
| "loss": 0.1016, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.5263157894736842, |
| "grad_norm": 1.5667542219161987, |
| "learning_rate": 1.905636093763031e-05, |
| "loss": 0.0951, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.6140350877192983, |
| "grad_norm": 1.686961054801941, |
| "learning_rate": 1.861226243180201e-05, |
| "loss": 0.1014, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.7017543859649122, |
| "grad_norm": 1.6863298416137695, |
| "learning_rate": 1.8090169943749477e-05, |
| "loss": 0.0891, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.7894736842105263, |
| "grad_norm": 1.7170166969299316, |
| "learning_rate": 1.7494811623892543e-05, |
| "loss": 0.1078, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.8771929824561403, |
| "grad_norm": 1.8812705278396606, |
| "learning_rate": 1.6831579129369347e-05, |
| "loss": 0.0984, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.9649122807017544, |
| "grad_norm": 2.126586437225342, |
| "learning_rate": 1.6106478796354382e-05, |
| "loss": 0.1002, |
| "step": 55 |
| }, |
| { |
| "epoch": 1.0526315789473684, |
| "grad_norm": 1.3081711530685425, |
| "learning_rate": 1.5326077245747998e-05, |
| "loss": 0.0872, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.1403508771929824, |
| "grad_norm": 1.7226194143295288, |
| "learning_rate": 1.449744191484066e-05, |
| "loss": 0.0621, |
| "step": 65 |
| }, |
| { |
| "epoch": 1.2280701754385965, |
| "grad_norm": 1.3189743757247925, |
| "learning_rate": 1.362807705350641e-05, |
| "loss": 0.0473, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.3157894736842106, |
| "grad_norm": 1.4051710367202759, |
| "learning_rate": 1.2725855764553981e-05, |
| "loss": 0.0542, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.4035087719298245, |
| "grad_norm": 1.6123563051223755, |
| "learning_rate": 1.179894870368854e-05, |
| "loss": 0.063, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.4912280701754386, |
| "grad_norm": 1.538110613822937, |
| "learning_rate": 1.08557500847884e-05, |
| "loss": 0.0484, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.5789473684210527, |
| "grad_norm": 1.4570614099502563, |
| "learning_rate": 9.904801660604234e-06, |
| "loss": 0.0495, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 1.2500959634780884, |
| "learning_rate": 8.954715367323468e-06, |
| "loss": 0.0466, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.7543859649122808, |
| "grad_norm": 1.505131721496582, |
| "learning_rate": 8.014095333542548e-06, |
| "loss": 0.0487, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.8421052631578947, |
| "grad_norm": 1.788224697113037, |
| "learning_rate": 7.0914599599460095e-06, |
| "loss": 0.0468, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.9298245614035088, |
| "grad_norm": 0.9474375247955322, |
| "learning_rate": 6.19516477535077e-06, |
| "loss": 0.0504, |
| "step": 110 |
| }, |
| { |
| "epoch": 2.017543859649123, |
| "grad_norm": 0.9824343919754028, |
| "learning_rate": 5.333326767743263e-06, |
| "loss": 0.0394, |
| "step": 115 |
| }, |
| { |
| "epoch": 2.1052631578947367, |
| "grad_norm": 0.7635042667388916, |
| "learning_rate": 4.513750875579303e-06, |
| "loss": 0.0264, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.192982456140351, |
| "grad_norm": 1.4104440212249756, |
| "learning_rate": 3.743859305052785e-06, |
| "loss": 0.0243, |
| "step": 125 |
| }, |
| { |
| "epoch": 2.280701754385965, |
| "grad_norm": 0.9276508688926697, |
| "learning_rate": 3.0306243134470668e-06, |
| "loss": 0.0208, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.3684210526315788, |
| "grad_norm": 0.7983176112174988, |
| "learning_rate": 2.380505067293293e-06, |
| "loss": 0.0169, |
| "step": 135 |
| }, |
| { |
| "epoch": 2.456140350877193, |
| "grad_norm": 0.9715234041213989, |
| "learning_rate": 1.7993891471580894e-06, |
| "loss": 0.023, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.543859649122807, |
| "grad_norm": 0.7858567833900452, |
| "learning_rate": 1.2925392288022299e-06, |
| "loss": 0.0206, |
| "step": 145 |
| }, |
| { |
| "epoch": 2.6315789473684212, |
| "grad_norm": 0.7981986999511719, |
| "learning_rate": 8.645454235739903e-07, |
| "loss": 0.0188, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.719298245614035, |
| "grad_norm": 0.6510503888130188, |
| "learning_rate": 5.192837096500058e-07, |
| "loss": 0.0184, |
| "step": 155 |
| }, |
| { |
| "epoch": 2.807017543859649, |
| "grad_norm": 0.9887875318527222, |
| "learning_rate": 2.5988083057666534e-07, |
| "loss": 0.0264, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.8947368421052633, |
| "grad_norm": 0.6808403730392456, |
| "learning_rate": 8.868597899638897e-08, |
| "loss": 0.0214, |
| "step": 165 |
| }, |
| { |
| "epoch": 2.982456140350877, |
| "grad_norm": 1.0281453132629395, |
| "learning_rate": 7.2495219952639636e-09, |
| "loss": 0.0178, |
| "step": 170 |
| }, |
| { |
| "epoch": 3.0, |
| "step": 171, |
| "total_flos": 0.0, |
| "train_loss": 0.05491512713474576, |
| "train_runtime": 12735.1501, |
| "train_samples_per_second": 0.43, |
| "train_steps_per_second": 0.013 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 171, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 8, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
