| { | |
| "best_metric": 1.7642425298690796, | |
| "best_model_checkpoint": "/net/tscratch/people/plgpiorczynskim/saved_models/relu-bert-base-uncased/checkpoint-5000", | |
| "epoch": 0.09378047865556306, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 4.975e-05, | |
| "loss": 2.2388, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 1.9998, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.9250000000000004e-05, | |
| "loss": 1.9622, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.9e-05, | |
| "loss": 1.9386, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.875e-05, | |
| "loss": 1.9249, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.85e-05, | |
| "loss": 1.914, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.825e-05, | |
| "loss": 1.9023, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.8955, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.775e-05, | |
| "loss": 1.8862, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.75e-05, | |
| "loss": 1.885, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_accuracy": 0.6519307229448076, | |
| "eval_loss": 1.7642425298690796, | |
| "eval_runtime": 2235.5625, | |
| "eval_samples_per_second": 961.542, | |
| "eval_steps_per_second": 1.252, | |
| "step": 5000 | |
| } | |
| ], | |
| "max_steps": 100000, | |
| "num_train_epochs": 2, | |
| "total_flos": 2.52676620288e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |