| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.65, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.016666666666666666, | |
| "eval_loss": 6.828747272491455, | |
| "eval_runtime": 3.1635, | |
| "eval_samples_per_second": 128.022, | |
| "eval_steps_per_second": 16.121, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "grad_norm": 2.298238754272461, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 6.479, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "eval_loss": 6.025122165679932, | |
| "eval_runtime": 3.1802, | |
| "eval_samples_per_second": 127.352, | |
| "eval_steps_per_second": 16.037, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 2.1337666511535645, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 5.4595, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "eval_loss": 5.005618572235107, | |
| "eval_runtime": 3.1314, | |
| "eval_samples_per_second": 129.334, | |
| "eval_steps_per_second": 16.287, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 2.1608963012695312, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 4.5889, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 4.342362403869629, | |
| "eval_runtime": 3.2423, | |
| "eval_samples_per_second": 124.91, | |
| "eval_steps_per_second": 15.729, | |
| "step": 30 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4279722779344896.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |