| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 4.0, |
| "eval_steps": 500, |
| "global_step": 2060, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.4854368932038835, |
| "grad_norm": 0.4750896096229553, |
| "learning_rate": 8.794946550048592e-05, |
| "loss": 0.854, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.970873786407767, |
| "grad_norm": 0.3962755501270294, |
| "learning_rate": 7.580174927113704e-05, |
| "loss": 0.4107, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.4563106796116505, |
| "grad_norm": 0.4067479968070984, |
| "learning_rate": 6.365403304178815e-05, |
| "loss": 0.3693, |
| "step": 750 |
| }, |
| { |
| "epoch": 1.941747572815534, |
| "grad_norm": 0.38044601678848267, |
| "learning_rate": 5.150631681243926e-05, |
| "loss": 0.3561, |
| "step": 1000 |
| }, |
| { |
| "epoch": 2.4271844660194173, |
| "grad_norm": 0.3728051483631134, |
| "learning_rate": 3.9358600583090386e-05, |
| "loss": 0.3473, |
| "step": 1250 |
| }, |
| { |
| "epoch": 2.912621359223301, |
| "grad_norm": 0.40094345808029175, |
| "learning_rate": 2.72108843537415e-05, |
| "loss": 0.3414, |
| "step": 1500 |
| }, |
| { |
| "epoch": 3.3980582524271843, |
| "grad_norm": 0.3950284719467163, |
| "learning_rate": 1.5063168124392615e-05, |
| "loss": 0.3321, |
| "step": 1750 |
| }, |
| { |
| "epoch": 3.883495145631068, |
| "grad_norm": 0.3837919235229492, |
| "learning_rate": 2.915451895043732e-06, |
| "loss": 0.3271, |
| "step": 2000 |
| }, |
| { |
| "epoch": 4.0, |
| "step": 2060, |
| "total_flos": 5.083160308163666e+17, |
| "train_loss": 0.4145523006476245, |
| "train_runtime": 7540.7405, |
| "train_samples_per_second": 69.933, |
| "train_steps_per_second": 0.273 |
| } |
| ], |
| "logging_steps": 250, |
| "max_steps": 2060, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": false, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 5.083160308163666e+17, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
