| { | |
| "best_metric": 0.8845644421803363, | |
| "best_model_checkpoint": "./outputs/electra-base-uncased/fine_tuned_models/checkpoint-36816", | |
| "epoch": 5.0, | |
| "global_step": 61360, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.7388593694651082e-05, | |
| "loss": 0.4468, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8792664289353032, | |
| "eval_loss": 0.32679134607315063, | |
| "eval_runtime": 24.6748, | |
| "eval_samples_per_second": 397.775, | |
| "eval_steps_per_second": 49.727, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.3041445270988312e-05, | |
| "loss": 0.2724, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8820173204279165, | |
| "eval_loss": 0.33915767073631287, | |
| "eval_runtime": 24.6385, | |
| "eval_samples_per_second": 398.36, | |
| "eval_steps_per_second": 49.8, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.694296847325541e-06, | |
| "loss": 0.1809, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8845644421803363, | |
| "eval_loss": 0.38803431391716003, | |
| "eval_runtime": 24.6189, | |
| "eval_samples_per_second": 398.678, | |
| "eval_steps_per_second": 49.84, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.3471484236627705e-06, | |
| "loss": 0.1219, | |
| "step": 49088 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8828323993886907, | |
| "eval_loss": 0.5000597834587097, | |
| "eval_runtime": 24.6471, | |
| "eval_samples_per_second": 398.222, | |
| "eval_steps_per_second": 49.783, | |
| "step": 49088 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0855, | |
| "step": 61360 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8833418237391747, | |
| "eval_loss": 0.5909910202026367, | |
| "eval_runtime": 24.6372, | |
| "eval_samples_per_second": 398.381, | |
| "eval_steps_per_second": 49.803, | |
| "step": 61360 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 61360, | |
| "total_flos": 1.6510003421181696e+17, | |
| "train_loss": 0.22150295809630155, | |
| "train_runtime": 13497.6716, | |
| "train_samples_per_second": 145.47, | |
| "train_steps_per_second": 4.546 | |
| } | |
| ], | |
| "max_steps": 61360, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.6510003421181696e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |