| { |
| "best_metric": 0.5064301619528891, |
| "best_model_checkpoint": "roberta-base/checkpoint-1929", |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 1929, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.78, |
| "grad_norm": 16.478424072265625, |
| "learning_rate": 4.742871954380509e-05, |
| "loss": 1.4921, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6762199845081333, |
| "eval_f1_macro": 0.36762215914350266, |
| "eval_f1_micro": 0.6762199845081333, |
| "eval_f1_weighted": 0.6381553142782734, |
| "eval_loss": 1.0307424068450928, |
| "eval_macro_fpr": 0.03508857639652692, |
| "eval_macro_sensitivity": 0.4027183655220987, |
| "eval_macro_specificity": 0.9746655126680946, |
| "eval_precision": 0.6361951106730956, |
| "eval_precision_macro": 0.3745780279088578, |
| "eval_recall": 0.6762199845081333, |
| "eval_recall_macro": 0.4027183655220987, |
| "eval_runtime": 65.0146, |
| "eval_samples_per_second": 19.857, |
| "eval_steps_per_second": 2.492, |
| "eval_weighted_fpr": 0.03306962025316456, |
| "eval_weighted_sensitivity": 0.6762199845081333, |
| "eval_weighted_specificity": 0.9437627055132857, |
| "step": 643 |
| }, |
| { |
| "epoch": 1.56, |
| "grad_norm": 11.052178382873535, |
| "learning_rate": 4.483670295489891e-05, |
| "loss": 0.9019, |
| "step": 1000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.737412858249419, |
| "eval_f1_macro": 0.4767493392755858, |
| "eval_f1_micro": 0.737412858249419, |
| "eval_f1_weighted": 0.7197263095885438, |
| "eval_loss": 0.8135831952095032, |
| "eval_macro_fpr": 0.025093207469991694, |
| "eval_macro_sensitivity": 0.5148741493607558, |
| "eval_macro_specificity": 0.9804978566683986, |
| "eval_precision": 0.7158327481302543, |
| "eval_precision_macro": 0.4625056436667857, |
| "eval_recall": 0.737412858249419, |
| "eval_recall_macro": 0.5148741493607558, |
| "eval_runtime": 78.3833, |
| "eval_samples_per_second": 16.47, |
| "eval_steps_per_second": 2.067, |
| "eval_weighted_fpr": 0.024804273066510573, |
| "eval_weighted_sensitivity": 0.737412858249419, |
| "eval_weighted_specificity": 0.9700549917765612, |
| "step": 1286 |
| }, |
| { |
| "epoch": 2.33, |
| "grad_norm": 16.875791549682617, |
| "learning_rate": 4.224468636599274e-05, |
| "loss": 0.7714, |
| "step": 1500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.7621998450813323, |
| "eval_f1_macro": 0.5064301619528891, |
| "eval_f1_micro": 0.7621998450813323, |
| "eval_f1_weighted": 0.7440787580787277, |
| "eval_loss": 0.8493772149085999, |
| "eval_macro_fpr": 0.022439395825223803, |
| "eval_macro_sensitivity": 0.5353612181578932, |
| "eval_macro_specificity": 0.9822327546765126, |
| "eval_precision": 0.7385639759374297, |
| "eval_precision_macro": 0.505572707108006, |
| "eval_recall": 0.7621998450813323, |
| "eval_recall_macro": 0.5353612181578932, |
| "eval_runtime": 62.8149, |
| "eval_samples_per_second": 20.552, |
| "eval_steps_per_second": 2.579, |
| "eval_weighted_fpr": 0.02179933252858056, |
| "eval_weighted_sensitivity": 0.7621998450813323, |
| "eval_weighted_specificity": 0.9712914750663563, |
| "step": 1929 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 9645, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 15, |
| "save_steps": 500, |
| "total_flos": 4056067183343616.0, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
