| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.2679738562091503, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.49378714208535923, | |
| "eval_f1_sklearn": 0.9530816971778758, | |
| "eval_loss": 0.12548573315143585, | |
| "eval_precision": 0.5784810126582278, | |
| "eval_precision_sklearn": 0.9530816971778758, | |
| "eval_recall": 0.43072573044297835, | |
| "eval_recall_sklearn": 0.9530816971778758, | |
| "eval_runtime": 4.9119, | |
| "eval_samples_per_second": 83.063, | |
| "eval_steps_per_second": 20.766, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.5957261229829918, | |
| "eval_f1_sklearn": 0.9530036301182716, | |
| "eval_loss": 0.11604497581720352, | |
| "eval_precision": 0.5543831168831169, | |
| "eval_precision_sklearn": 0.9530036301182716, | |
| "eval_recall": 0.6437323279924599, | |
| "eval_recall_sklearn": 0.9530036301182716, | |
| "eval_runtime": 4.3163, | |
| "eval_samples_per_second": 94.524, | |
| "eval_steps_per_second": 23.631, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.5958316305680426, | |
| "eval_f1_sklearn": 0.9508958195089582, | |
| "eval_loss": 0.14478950202465057, | |
| "eval_precision": 0.525974025974026, | |
| "eval_precision_sklearn": 0.9508958195089582, | |
| "eval_recall": 0.6870876531573987, | |
| "eval_recall_sklearn": 0.9508958195089582, | |
| "eval_runtime": 4.3139, | |
| "eval_samples_per_second": 94.578, | |
| "eval_steps_per_second": 23.645, | |
| "step": 459 | |
| }, | |
| { | |
| "epoch": 3.2679738562091503, | |
| "grad_norm": 0.4012909531593323, | |
| "learning_rate": 3.464052287581699e-05, | |
| "loss": 0.0651, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 765, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 433135952304768.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 0.0001, | |
| "metric": "eval/loss", | |
| "per_device_train_batch_size": 8 | |
| } | |
| } | |