| { | |
| "best_metric": 0.7524875621890547, | |
| "best_model_checkpoint": "./outputs/deberta-base-output/vua_metaphor_32_0.1_0.00005_03-21-23_21-55/checkpoint-906", | |
| "epoch": 4.0, | |
| "global_step": 1208, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.9911894273127756e-05, | |
| "loss": 0.1588, | |
| "step": 302 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9615206305053314, | |
| "eval_f1": 0.6345506345506345, | |
| "eval_loss": 0.09228184819221497, | |
| "eval_precision": 0.8226997985224983, | |
| "eval_recall": 0.5164418212478921, | |
| "eval_runtime": 7.3935, | |
| "eval_samples_per_second": 325.962, | |
| "eval_steps_per_second": 40.847, | |
| "step": 302 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2.6607929515418507e-05, | |
| "loss": 0.0661, | |
| "step": 604 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9673020807766779, | |
| "eval_f1": 0.7279328341275245, | |
| "eval_loss": 0.08415527641773224, | |
| "eval_precision": 0.7882063882063882, | |
| "eval_recall": 0.6762225969645869, | |
| "eval_runtime": 7.0018, | |
| "eval_samples_per_second": 344.196, | |
| "eval_steps_per_second": 43.132, | |
| "step": 604 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.3303964757709254e-05, | |
| "loss": 0.0358, | |
| "step": 906 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9674384357359077, | |
| "eval_f1": 0.7524875621890547, | |
| "eval_loss": 0.10287901014089584, | |
| "eval_precision": 0.7402120717781403, | |
| "eval_recall": 0.765177065767285, | |
| "eval_runtime": 6.9543, | |
| "eval_samples_per_second": 346.549, | |
| "eval_steps_per_second": 43.426, | |
| "step": 906 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.019, | |
| "step": 1208 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.968229294499441, | |
| "eval_f1": 0.738378621154278, | |
| "eval_loss": 0.11687722057104111, | |
| "eval_precision": 0.7900048053820279, | |
| "eval_recall": 0.6930860033726813, | |
| "eval_runtime": 6.91, | |
| "eval_samples_per_second": 348.768, | |
| "eval_steps_per_second": 43.704, | |
| "step": 1208 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 1208, | |
| "total_flos": 1154456176071726.0, | |
| "train_loss": 0.0699166256860392, | |
| "train_runtime": 336.8243, | |
| "train_samples_per_second": 114.398, | |
| "train_steps_per_second": 3.586 | |
| } | |
| ], | |
| "max_steps": 1208, | |
| "num_train_epochs": 4, | |
| "total_flos": 1154456176071726.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |