| { | |
| "best_metric": 0.8635321100917431, | |
| "best_model_checkpoint": "bert-base-uncased-finetuned-sst2/run-2/checkpoint-1684", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 1684, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 267.4592590332031, | |
| "learning_rate": 6.490431462154475e-05, | |
| "loss": 0.6127, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 9.022435188293457, | |
| "learning_rate": 3.749539797393294e-05, | |
| "loss": 0.5862, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 10.805120468139648, | |
| "learning_rate": 1.0086481326321143e-05, | |
| "loss": 0.4922, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8635321100917431, | |
| "eval_loss": 0.47688528895378113, | |
| "eval_runtime": 2.4, | |
| "eval_samples_per_second": 363.329, | |
| "eval_steps_per_second": 22.916, | |
| "step": 1684 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1684, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 73089373401840.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 9.231323126915654e-05, | |
| "num_train_epochs": 1, | |
| "per_device_train_batch_size": 4, | |
| "seed": 29 | |
| } | |
| } | |