| { | |
| "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": 0.515832482124617, | |
| "eval_f1_sklearn": 0.9513251883367813, | |
| "eval_loss": 0.12229713052511215, | |
| "eval_precision": 0.5629877369007804, | |
| "eval_precision_sklearn": 0.9513251883367813, | |
| "eval_recall": 0.4759660697455231, | |
| "eval_recall_sklearn": 0.9513251883367813, | |
| "eval_runtime": 4.2613, | |
| "eval_samples_per_second": 95.745, | |
| "eval_steps_per_second": 23.936, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 1.6339869281045751, | |
| "grad_norm": 1.7371214628219604, | |
| "learning_rate": 6.73202614379085e-05, | |
| "loss": 0.1032, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.507920792079208, | |
| "eval_f1_sklearn": 0.9493735118466763, | |
| "eval_loss": 0.16793084144592285, | |
| "eval_precision": 0.5349322210636079, | |
| "eval_precision_sklearn": 0.9493735118466763, | |
| "eval_recall": 0.48350612629594725, | |
| "eval_recall_sklearn": 0.9493735118466763, | |
| "eval_runtime": 6.1428, | |
| "eval_samples_per_second": 66.42, | |
| "eval_steps_per_second": 16.605, | |
| "step": 612 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.6094600251151109, | |
| "eval_f1_sklearn": 0.9524181271712401, | |
| "eval_loss": 0.16153906285762787, | |
| "eval_precision": 0.5481927710843374, | |
| "eval_precision_sklearn": 0.9524181271712401, | |
| "eval_recall": 0.6861451460885957, | |
| "eval_recall_sklearn": 0.9524181271712401, | |
| "eval_runtime": 5.1328, | |
| "eval_samples_per_second": 79.488, | |
| "eval_steps_per_second": 19.872, | |
| "step": 918 | |
| }, | |
| { | |
| "epoch": 3.2679738562091503, | |
| "grad_norm": 0.30743545293807983, | |
| "learning_rate": 3.464052287581699e-05, | |
| "loss": 0.0309, | |
| "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": 0.0001, | |
| "metric": "eval/loss", | |
| "per_device_train_batch_size": 4 | |
| } | |
| } | |