| { | |
| "best_metric": 0.4764873250102919, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-6/checkpoint-3207", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 3207, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 5.3938263247342015e-06, | |
| "loss": 0.5711, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.397552529992015e-06, | |
| "loss": 0.5074, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5285322070121765, | |
| "eval_matthews_correlation": 0.3822545955758537, | |
| "eval_runtime": 0.7108, | |
| "eval_samples_per_second": 1467.388, | |
| "eval_steps_per_second": 92.855, | |
| "step": 1069 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 3.401278735249827e-06, | |
| "loss": 0.4345, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 2.40500494050764e-06, | |
| "loss": 0.4083, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5192927122116089, | |
| "eval_matthews_correlation": 0.4700719675330965, | |
| "eval_runtime": 0.7849, | |
| "eval_samples_per_second": 1328.766, | |
| "eval_steps_per_second": 84.083, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 1.4087311457654528e-06, | |
| "loss": 0.3741, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 4.1245735102326556e-07, | |
| "loss": 0.3494, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.5613988041877747, | |
| "eval_matthews_correlation": 0.4764873250102919, | |
| "eval_runtime": 0.7007, | |
| "eval_samples_per_second": 1488.508, | |
| "eval_steps_per_second": 94.191, | |
| "step": 3207 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3207, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 113311112203092.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 6.390100119476389e-06, | |
| "num_train_epochs": 3, | |
| "per_device_train_batch_size": 8, | |
| "seed": 30 | |
| } | |
| } | |