| { | |
| "best_metric": 0.622779879885066, | |
| "best_model_checkpoint": "students_scores_model/run-0/checkpoint-1689", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 1689, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8880994671403197, | |
| "grad_norm": 9.486105918884277, | |
| "learning_rate": 1.6060239187203855e-05, | |
| "loss": 1.0687, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.559496193541961, | |
| "eval_loss": 1.0184900760650635, | |
| "eval_runtime": 37.7502, | |
| "eval_samples_per_second": 59.602, | |
| "eval_steps_per_second": 7.47, | |
| "step": 563 | |
| }, | |
| { | |
| "epoch": 1.7761989342806395, | |
| "grad_norm": 16.897672653198242, | |
| "learning_rate": 9.306564171558836e-06, | |
| "loss": 0.8626, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.6141765793522762, | |
| "eval_loss": 0.877347469329834, | |
| "eval_runtime": 37.6032, | |
| "eval_samples_per_second": 59.835, | |
| "eval_steps_per_second": 7.499, | |
| "step": 1126 | |
| }, | |
| { | |
| "epoch": 2.664298401420959, | |
| "grad_norm": 14.929506301879883, | |
| "learning_rate": 2.552889155913817e-06, | |
| "loss": 0.7589, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.622779879885066, | |
| "eval_loss": 0.8738921880722046, | |
| "eval_runtime": 38.6362, | |
| "eval_samples_per_second": 58.236, | |
| "eval_steps_per_second": 7.299, | |
| "step": 1689 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1689, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6310155946881024.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 2.2813914202848873e-05, | |
| "num_train_epochs": 3, | |
| "per_device_train_batch_size": 16, | |
| "weight_decay": 0.0012064342197859815 | |
| } | |
| } | |