| { | |
| "best_metric": 0.11405563354492188, | |
| "best_model_checkpoint": "BERT-HARD-balanced/checkpoint-1500", | |
| "epoch": 1.680672268907563, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.2997198879551826e-05, | |
| "loss": 0.1682, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_accuracy": 0.950212866603595, | |
| "eval_f1": 0.9498032669607726, | |
| "eval_loss": 0.13225148618221283, | |
| "eval_precision": 0.9576821351286366, | |
| "eval_recall": 0.9420529801324503, | |
| "eval_runtime": 6.9677, | |
| "eval_samples_per_second": 1213.595, | |
| "eval_steps_per_second": 19.088, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.5994397759103643e-05, | |
| "loss": 0.1337, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_accuracy": 0.9571901608325449, | |
| "eval_f1": 0.9578481602235678, | |
| "eval_loss": 0.12283609807491302, | |
| "eval_precision": 0.943348623853211, | |
| "eval_recall": 0.9728003784295175, | |
| "eval_runtime": 6.9706, | |
| "eval_samples_per_second": 1213.091, | |
| "eval_steps_per_second": 19.08, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 2.8991596638655467e-05, | |
| "loss": 0.1132, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_accuracy": 0.9605014191106906, | |
| "eval_f1": 0.9605294256676908, | |
| "eval_loss": 0.11405563354492188, | |
| "eval_precision": 0.9598488427019367, | |
| "eval_recall": 0.9612109744560076, | |
| "eval_runtime": 7.0035, | |
| "eval_samples_per_second": 1207.4, | |
| "eval_steps_per_second": 18.991, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 2.1988795518207285e-05, | |
| "loss": 0.0999, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_accuracy": 0.9609744560075686, | |
| "eval_f1": 0.9610940815845319, | |
| "eval_loss": 0.1178417056798935, | |
| "eval_precision": 0.9581570286788904, | |
| "eval_recall": 0.9640491958372753, | |
| "eval_runtime": 6.9707, | |
| "eval_samples_per_second": 1213.078, | |
| "eval_steps_per_second": 19.08, | |
| "step": 2000 | |
| } | |
| ], | |
| "max_steps": 3570, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.4904145310863872e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |