| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.2679738562091503, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.381578947368421, | |
| "eval_f1_sklearn": 0.9460166282836957, | |
| "eval_loss": 0.1464981883764267, | |
| "eval_precision": 0.45609436435124506, | |
| "eval_precision_sklearn": 0.9460166282836957, | |
| "eval_recall": 0.32799245994344955, | |
| "eval_recall_sklearn": 0.9460166282836957, | |
| "eval_runtime": 4.2583, | |
| "eval_samples_per_second": 95.813, | |
| "eval_steps_per_second": 23.953, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 1.6339869281045751, | |
| "grad_norm": 5.414714813232422, | |
| "learning_rate": 3.3660130718954253e-06, | |
| "loss": 0.1668, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.4384873145045477, | |
| "eval_f1_sklearn": 0.9487099418400406, | |
| "eval_loss": 0.1396203637123108, | |
| "eval_precision": 0.4455252918287938, | |
| "eval_precision_sklearn": 0.9487099418400406, | |
| "eval_recall": 0.43166823751178135, | |
| "eval_recall_sklearn": 0.9487099418400406, | |
| "eval_runtime": 5.2684, | |
| "eval_samples_per_second": 77.443, | |
| "eval_steps_per_second": 19.361, | |
| "step": 612 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.49019607843137253, | |
| "eval_f1_sklearn": 0.9503103165619267, | |
| "eval_loss": 0.14265619218349457, | |
| "eval_precision": 0.46491969568892644, | |
| "eval_precision_sklearn": 0.9503103165619267, | |
| "eval_recall": 0.5183788878416589, | |
| "eval_recall_sklearn": 0.9503103165619267, | |
| "eval_runtime": 5.7901, | |
| "eval_samples_per_second": 70.465, | |
| "eval_steps_per_second": 17.616, | |
| "step": 918 | |
| }, | |
| { | |
| "epoch": 3.2679738562091503, | |
| "grad_norm": 4.938503742218018, | |
| "learning_rate": 1.7320261437908499e-06, | |
| "loss": 0.0872, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1530, | |
| "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": 318346024442760.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 5e-06, | |
| "metric": "eval/loss", | |
| "per_device_train_batch_size": 4 | |
| } | |
| } | |