| { | |
| "best_metric": 0.5011003042513985, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-1/checkpoint-4276", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 4276, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 7.5246806144714355, | |
| "learning_rate": 1.1577749465690007e-05, | |
| "loss": 0.5887, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 25.256967544555664, | |
| "learning_rate": 1.0044678826695038e-05, | |
| "loss": 0.5585, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 10.809298515319824, | |
| "learning_rate": 8.511608187700069e-06, | |
| "loss": 0.5535, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 55.36846160888672, | |
| "learning_rate": 6.978537548705099e-06, | |
| "loss": 0.5117, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6122316122055054, | |
| "eval_matthews_correlation": 0.45831111658186324, | |
| "eval_runtime": 0.8138, | |
| "eval_samples_per_second": 1281.66, | |
| "eval_steps_per_second": 81.102, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 71.29000854492188, | |
| "learning_rate": 5.44546690971013e-06, | |
| "loss": 0.4209, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 3.4101831912994385, | |
| "learning_rate": 3.912396270715161e-06, | |
| "loss": 0.4658, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 2.0748186111450195, | |
| "learning_rate": 2.379325631720192e-06, | |
| "loss": 0.4685, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 153.2447052001953, | |
| "learning_rate": 8.46254992725223e-07, | |
| "loss": 0.4715, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.7943716645240784, | |
| "eval_matthews_correlation": 0.5011003042513985, | |
| "eval_runtime": 0.7454, | |
| "eval_samples_per_second": 1399.229, | |
| "eval_steps_per_second": 88.542, | |
| "step": 4276 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4276, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "total_flos": 65361794770440.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.3110820104684976e-05, | |
| "num_train_epochs": 2, | |
| "per_device_train_batch_size": 4, | |
| "seed": 22 | |
| } | |
| } | |