| { | |
| "best_metric": 0.91, | |
| "best_model_checkpoint": "gemma-finetuned-spam/run-1/checkpoint-116", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 116, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.82, | |
| "eval_loss": 0.4550154209136963, | |
| "eval_runtime": 4.9133, | |
| "eval_samples_per_second": 20.353, | |
| "eval_steps_per_second": 10.176, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.88, | |
| "eval_loss": 0.33550891280174255, | |
| "eval_runtime": 5.1025, | |
| "eval_samples_per_second": 19.598, | |
| "eval_steps_per_second": 9.799, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9, | |
| "eval_loss": 0.2563902735710144, | |
| "eval_runtime": 5.0535, | |
| "eval_samples_per_second": 19.788, | |
| "eval_steps_per_second": 9.894, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.4482758620689653, | |
| "grad_norm": 2.428945779800415, | |
| "learning_rate": 3.566130042186149e-06, | |
| "loss": 0.3325, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.91, | |
| "eval_loss": 0.23370271921157837, | |
| "eval_runtime": 4.8953, | |
| "eval_samples_per_second": 20.428, | |
| "eval_steps_per_second": 10.214, | |
| "step": 116 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 116, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 1, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.9245204812857344e+16, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 2.585444280584958e-05, | |
| "num_train_epochs": 4, | |
| "per_device_train_batch_size": 32, | |
| "seed": 5 | |
| } | |
| } | |