| { |
| "best_metric": 0.665659747836216, |
| "best_model_checkpoint": "roberta-base/checkpoint-4501", |
| "epoch": 7.0, |
| "eval_steps": 500, |
| "global_step": 4501, |
| "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 |
| }, |
| { |
| "epoch": 3.11, |
| "grad_norm": 59.48842239379883, |
| "learning_rate": 3.9652669777086575e-05, |
| "loss": 0.7018, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.89, |
| "grad_norm": 1.6855363845825195, |
| "learning_rate": 3.706583722135822e-05, |
| "loss": 0.5351, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.7645236250968241, |
| "eval_f1_macro": 0.544445143585482, |
| "eval_f1_micro": 0.7645236250968241, |
| "eval_f1_weighted": 0.7545581478665218, |
| "eval_loss": 0.9904704093933105, |
| "eval_macro_fpr": 0.022368675533492623, |
| "eval_macro_sensitivity": 0.5653277909563001, |
| "eval_macro_specificity": 0.9821237739107975, |
| "eval_precision": 0.7656722240785093, |
| "eval_precision_macro": 0.5495765557734748, |
| "eval_recall": 0.7645236250968241, |
| "eval_recall_macro": 0.5653277909563001, |
| "eval_runtime": 80.7211, |
| "eval_samples_per_second": 15.993, |
| "eval_steps_per_second": 2.007, |
| "eval_weighted_fpr": 0.021526695935419914, |
| "eval_weighted_sensitivity": 0.7645236250968241, |
| "eval_weighted_specificity": 0.9673329835651372, |
| "step": 2572 |
| }, |
| { |
| "epoch": 4.67, |
| "grad_norm": 41.18514633178711, |
| "learning_rate": 3.447382063245205e-05, |
| "loss": 0.4521, |
| "step": 3000 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.7900852052672347, |
| "eval_f1_macro": 0.6116525431724917, |
| "eval_f1_micro": 0.7900852052672347, |
| "eval_f1_weighted": 0.7794301250682801, |
| "eval_loss": 1.0259578227996826, |
| "eval_macro_fpr": 0.019043445988564375, |
| "eval_macro_sensitivity": 0.6315170743424566, |
| "eval_macro_specificity": 0.984305159354804, |
| "eval_precision": 0.7882093327742286, |
| "eval_precision_macro": 0.6420940275483907, |
| "eval_recall": 0.7900852052672347, |
| "eval_recall_macro": 0.6315170743424566, |
| "eval_runtime": 56.9964, |
| "eval_samples_per_second": 22.651, |
| "eval_steps_per_second": 2.842, |
| "eval_weighted_fpr": 0.018624149542986735, |
| "eval_weighted_sensitivity": 0.7900852052672347, |
| "eval_weighted_specificity": 0.9744921850548264, |
| "step": 3215 |
| }, |
| { |
| "epoch": 5.44, |
| "grad_norm": 92.02596282958984, |
| "learning_rate": 3.1881804043545884e-05, |
| "loss": 0.3466, |
| "step": 3500 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.7869868319132456, |
| "eval_f1_macro": 0.6118575336738351, |
| "eval_f1_micro": 0.7869868319132455, |
| "eval_f1_weighted": 0.7889880056842663, |
| "eval_loss": 1.0384949445724487, |
| "eval_macro_fpr": 0.019220806323432427, |
| "eval_macro_sensitivity": 0.633518783056355, |
| "eval_macro_specificity": 0.984312476194111, |
| "eval_precision": 0.8134105657387156, |
| "eval_precision_macro": 0.6722188161248251, |
| "eval_recall": 0.7869868319132456, |
| "eval_recall_macro": 0.633518783056355, |
| "eval_runtime": 58.1615, |
| "eval_samples_per_second": 22.197, |
| "eval_steps_per_second": 2.785, |
| "eval_weighted_fpr": 0.018966825298296436, |
| "eval_weighted_sensitivity": 0.7869868319132456, |
| "eval_weighted_specificity": 0.9777003109984174, |
| "step": 3858 |
| }, |
| { |
| "epoch": 6.22, |
| "grad_norm": 80.3629379272461, |
| "learning_rate": 2.9294971487817523e-05, |
| "loss": 0.3321, |
| "step": 4000 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 0.3377046585083008, |
| "learning_rate": 2.6702954898911353e-05, |
| "loss": 0.2333, |
| "step": 4500 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8117738187451587, |
| "eval_f1_macro": 0.665659747836216, |
| "eval_f1_micro": 0.8117738187451587, |
| "eval_f1_weighted": 0.8057624075524522, |
| "eval_loss": 1.1464879512786865, |
| "eval_macro_fpr": 0.01693769060060091, |
| "eval_macro_sensitivity": 0.6784424177464591, |
| "eval_macro_specificity": 0.9857572234742, |
| "eval_precision": 0.8064435087918365, |
| "eval_precision_macro": 0.6794791733053616, |
| "eval_recall": 0.8117738187451587, |
| "eval_recall_macro": 0.6784424177464591, |
| "eval_runtime": 54.6308, |
| "eval_samples_per_second": 23.631, |
| "eval_steps_per_second": 2.965, |
| "eval_weighted_fpr": 0.016292323164599398, |
| "eval_weighted_sensitivity": 0.8117738187451587, |
| "eval_weighted_specificity": 0.9745845333678407, |
| "step": 4501 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 9645, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 15, |
| "save_steps": 500, |
| "total_flos": 9464156761135104.0, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
