| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.005142589995325, | |
| "eval_steps": 27, | |
| "global_step": 134, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0074801309022907905, | |
| "eval_loss": 4.825988292694092, | |
| "eval_runtime": 15.2212, | |
| "eval_samples_per_second": 14.848, | |
| "eval_steps_per_second": 7.424, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.20196353436185133, | |
| "eval_loss": 4.658951759338379, | |
| "eval_runtime": 15.2507, | |
| "eval_samples_per_second": 14.819, | |
| "eval_steps_per_second": 7.409, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2992052360916316, | |
| "grad_norm": 59.90390396118164, | |
| "learning_rate": 1e-05, | |
| "loss": 75.3596, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.40392706872370265, | |
| "eval_loss": 3.6523022651672363, | |
| "eval_runtime": 14.9676, | |
| "eval_samples_per_second": 15.099, | |
| "eval_steps_per_second": 7.55, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.5984104721832632, | |
| "grad_norm": 43.99336624145508, | |
| "learning_rate": 2e-05, | |
| "loss": 57.7534, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.605890603085554, | |
| "eval_loss": 3.0761516094207764, | |
| "eval_runtime": 14.9857, | |
| "eval_samples_per_second": 15.081, | |
| "eval_steps_per_second": 7.541, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.8078541374474053, | |
| "eval_loss": 2.918757438659668, | |
| "eval_runtime": 15.1858, | |
| "eval_samples_per_second": 14.882, | |
| "eval_steps_per_second": 7.441, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.8976157082748948, | |
| "grad_norm": 46.91463088989258, | |
| "learning_rate": 3.1375836213126653e-06, | |
| "loss": 48.2037, | |
| "step": 120 | |
| } | |
| ], | |
| "logging_steps": 40, | |
| "max_steps": 134, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 27, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.394310052682138e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |