| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0008270443502532823, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.6540887005065647e-05, | |
| "eval_loss": 10.818584442138672, | |
| "eval_runtime": 178.2631, | |
| "eval_samples_per_second": 142.8, | |
| "eval_steps_per_second": 71.4, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00016540887005065647, | |
| "grad_norm": 5.854555606842041, | |
| "learning_rate": 4e-05, | |
| "loss": 10.7327, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0002150315310658534, | |
| "eval_loss": 10.767061233520508, | |
| "eval_runtime": 179.0502, | |
| "eval_samples_per_second": 142.172, | |
| "eval_steps_per_second": 71.086, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00033081774010131293, | |
| "grad_norm": 3.983314037322998, | |
| "learning_rate": 8e-05, | |
| "loss": 10.7773, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0004300630621317068, | |
| "eval_loss": 10.505905151367188, | |
| "eval_runtime": 179.6613, | |
| "eval_samples_per_second": 141.689, | |
| "eval_steps_per_second": 70.844, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0004962266101519694, | |
| "grad_norm": 4.059611797332764, | |
| "learning_rate": 0.00012, | |
| "loss": 10.5995, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0006450945931975602, | |
| "eval_loss": 10.089973449707031, | |
| "eval_runtime": 178.4401, | |
| "eval_samples_per_second": 142.658, | |
| "eval_steps_per_second": 71.329, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0006616354802026259, | |
| "grad_norm": 3.5047848224639893, | |
| "learning_rate": 0.00016, | |
| "loss": 10.116, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0008270443502532823, | |
| "grad_norm": 4.182229995727539, | |
| "learning_rate": 0.0002, | |
| "loss": 9.9025, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4982449373184.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |