| { | |
| "best_metric": 0.692, | |
| "best_model_checkpoint": "nli_results/10k_test3/models/bert/checkpoint-625", | |
| "epoch": 2.9952, | |
| "global_step": 936, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 4.99465811965812e-05, | |
| "loss": 1.0528, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.465811965811966e-05, | |
| "loss": 0.9967, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.931623931623932e-05, | |
| "loss": 0.9252, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 3.402777777777778e-05, | |
| "loss": 0.8753, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5313333333333333, | |
| "eval_f1_weighted": 0.5138275157101065, | |
| "eval_loss": 0.9772958755493164, | |
| "eval_precision_weighted": 0.627640816674217, | |
| "eval_recall_weighted": 0.5313333333333333, | |
| "eval_runtime": 3.4984, | |
| "eval_samples_per_second": 428.762, | |
| "eval_steps_per_second": 107.19, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 2.8685897435897437e-05, | |
| "loss": 0.7221, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 2.3344017094017095e-05, | |
| "loss": 0.6875, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.8002136752136753e-05, | |
| "loss": 0.6746, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.692, | |
| "eval_f1_weighted": 0.6927478366098441, | |
| "eval_loss": 0.7383598685264587, | |
| "eval_precision_weighted": 0.695395361284498, | |
| "eval_recall_weighted": 0.692, | |
| "eval_runtime": 3.4271, | |
| "eval_samples_per_second": 437.694, | |
| "eval_steps_per_second": 109.424, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 1.2660256410256411e-05, | |
| "loss": 0.4918, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 7.371794871794873e-06, | |
| "loss": 0.4254, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 2.02991452991453e-06, | |
| "loss": 0.4239, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6693333333333333, | |
| "eval_f1_weighted": 0.6689803827787892, | |
| "eval_loss": 0.907555103302002, | |
| "eval_precision_weighted": 0.6888550655655988, | |
| "eval_recall_weighted": 0.6693333333333333, | |
| "eval_runtime": 3.3249, | |
| "eval_samples_per_second": 451.136, | |
| "eval_steps_per_second": 112.784, | |
| "step": 936 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 936, | |
| "total_flos": 1158348599253216.0, | |
| "train_loss": 0.6807532799549592, | |
| "train_runtime": 369.8644, | |
| "train_samples_per_second": 81.111, | |
| "train_steps_per_second": 2.531 | |
| } | |
| ], | |
| "max_steps": 936, | |
| "num_train_epochs": 3, | |
| "total_flos": 1158348599253216.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |