| { |
| "best_global_step": 3824, |
| "best_metric": 0.7665036674816625, |
| "best_model_checkpoint": "./phobert-multitask/checkpoint-3824", |
| "epoch": 8.0, |
| "eval_steps": 500, |
| "global_step": 3824, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 6.267406940460205, |
| "learning_rate": 9.981598232331348e-06, |
| "loss": 2.043239577544783, |
| "step": 478 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy_coarse": 0.7426650366748166, |
| "eval_accuracy_fine": 0.628361858190709, |
| "eval_f1_coarse_macro": 0.6502651710386934, |
| "eval_f1_fine_macro": 0.605379593432809, |
| "eval_f1_fine_weighted": 0.5995232314641942, |
| "eval_loss": 1.3452860116958618, |
| "eval_runtime": 18.5356, |
| "eval_samples_per_second": 88.262, |
| "eval_steps_per_second": 5.557, |
| "step": 478 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 24.215755462646484, |
| "learning_rate": 9.33354666334534e-06, |
| "loss": 1.203696534224634, |
| "step": 956 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy_coarse": 0.8135696821515892, |
| "eval_accuracy_fine": 0.7408312958435208, |
| "eval_f1_coarse_macro": 0.802456160930916, |
| "eval_f1_fine_macro": 0.751035231453591, |
| "eval_f1_fine_weighted": 0.744689454665699, |
| "eval_loss": 1.0884628295898438, |
| "eval_runtime": 18.5112, |
| "eval_samples_per_second": 88.379, |
| "eval_steps_per_second": 5.564, |
| "step": 956 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.8972363471984863, |
| "learning_rate": 7.873177237961524e-06, |
| "loss": 0.9721315774957505, |
| "step": 1434 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy_coarse": 0.8166259168704156, |
| "eval_accuracy_fine": 0.7353300733496333, |
| "eval_f1_coarse_macro": 0.7919295893473026, |
| "eval_f1_fine_macro": 0.7394243754914572, |
| "eval_f1_fine_weighted": 0.732021202826222, |
| "eval_loss": 1.0886324644088745, |
| "eval_runtime": 18.5493, |
| "eval_samples_per_second": 88.197, |
| "eval_steps_per_second": 5.553, |
| "step": 1434 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.9366881847381592, |
| "learning_rate": 5.8742343630904545e-06, |
| "loss": 0.8340984009299817, |
| "step": 1912 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy_coarse": 0.8257946210268948, |
| "eval_accuracy_fine": 0.7524449877750611, |
| "eval_f1_coarse_macro": 0.8115790760213039, |
| "eval_f1_fine_macro": 0.7585230889291659, |
| "eval_f1_fine_weighted": 0.7513882818781048, |
| "eval_loss": 1.079779028892517, |
| "eval_runtime": 18.5463, |
| "eval_samples_per_second": 88.212, |
| "eval_steps_per_second": 5.554, |
| "step": 1912 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 19.028635025024414, |
| "learning_rate": 3.7114173594782087e-06, |
| "loss": 0.7354573525145462, |
| "step": 2390 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy_coarse": 0.8257946210268948, |
| "eval_accuracy_fine": 0.7579462102689487, |
| "eval_f1_coarse_macro": 0.8118641202186286, |
| "eval_f1_fine_macro": 0.7671818296330676, |
| "eval_f1_fine_weighted": 0.7596953624722936, |
| "eval_loss": 1.1095541715621948, |
| "eval_runtime": 18.5366, |
| "eval_samples_per_second": 88.258, |
| "eval_steps_per_second": 5.557, |
| "step": 2390 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 0.5821412205696106, |
| "learning_rate": 1.7901435466186506e-06, |
| "loss": 0.6709438308013533, |
| "step": 2868 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy_coarse": 0.8221271393643031, |
| "eval_accuracy_fine": 0.7646699266503667, |
| "eval_f1_coarse_macro": 0.8110735990814336, |
| "eval_f1_fine_macro": 0.7724096547525416, |
| "eval_f1_fine_weighted": 0.7649762626590634, |
| "eval_loss": 1.1057395935058594, |
| "eval_runtime": 18.5515, |
| "eval_samples_per_second": 88.187, |
| "eval_steps_per_second": 5.552, |
| "step": 2868 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 6.050240516662598, |
| "learning_rate": 4.705532775431976e-07, |
| "loss": 0.6313777269179851, |
| "step": 3346 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy_coarse": 0.8227383863080685, |
| "eval_accuracy_fine": 0.7652811735941321, |
| "eval_f1_coarse_macro": 0.8124455379854854, |
| "eval_f1_fine_macro": 0.7740002317386471, |
| "eval_f1_fine_weighted": 0.7668867937841447, |
| "eval_loss": 1.120171308517456, |
| "eval_runtime": 18.6422, |
| "eval_samples_per_second": 87.758, |
| "eval_steps_per_second": 5.525, |
| "step": 3346 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 3.5955278873443604, |
| "learning_rate": 2.0838688302715625e-12, |
| "loss": 0.6137639129510983, |
| "step": 3824 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy_coarse": 0.8227383863080685, |
| "eval_accuracy_fine": 0.7665036674816625, |
| "eval_f1_coarse_macro": 0.8111588897237633, |
| "eval_f1_fine_macro": 0.7752350873790258, |
| "eval_f1_fine_weighted": 0.7681455373546499, |
| "eval_loss": 1.12169349193573, |
| "eval_runtime": 18.581, |
| "eval_samples_per_second": 88.047, |
| "eval_steps_per_second": 5.543, |
| "step": 3824 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 3824, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 8, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 3, |
| "early_stopping_threshold": 0.0 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 0 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
