| { |
| "best_global_step": 60, |
| "best_metric": 0.5803486108779907, |
| "best_model_checkpoint": "tiny_bert_km_20_v1_mrpc/checkpoint-60", |
| "epoch": 9.0, |
| "eval_steps": 500, |
| "global_step": 135, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.7678272128105164, |
| "learning_rate": 4.9e-05, |
| "loss": 0.6272, |
| "step": 15 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.696078431372549, |
| "eval_combined_score": 0.7517659237608086, |
| "eval_f1": 0.8074534161490683, |
| "eval_loss": 0.5963134765625, |
| "eval_runtime": 0.1363, |
| "eval_samples_per_second": 2992.343, |
| "eval_steps_per_second": 14.668, |
| "step": 15 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.2829147577285767, |
| "learning_rate": 4.8e-05, |
| "loss": 0.5912, |
| "step": 30 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.696078431372549, |
| "eval_combined_score": 0.7560510851224762, |
| "eval_f1": 0.8160237388724035, |
| "eval_loss": 0.599238395690918, |
| "eval_runtime": 0.1408, |
| "eval_samples_per_second": 2898.125, |
| "eval_steps_per_second": 14.206, |
| "step": 30 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.9696577787399292, |
| "learning_rate": 4.7e-05, |
| "loss": 0.5662, |
| "step": 45 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.7156862745098039, |
| "eval_combined_score": 0.7704937396645406, |
| "eval_f1": 0.8253012048192772, |
| "eval_loss": 0.5903894901275635, |
| "eval_runtime": 0.1431, |
| "eval_samples_per_second": 2851.713, |
| "eval_steps_per_second": 13.979, |
| "step": 45 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 1.1414984464645386, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.5452, |
| "step": 60 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.7156862745098039, |
| "eval_combined_score": 0.7680598555211559, |
| "eval_f1": 0.8204334365325078, |
| "eval_loss": 0.5803486108779907, |
| "eval_runtime": 0.1553, |
| "eval_samples_per_second": 2627.042, |
| "eval_steps_per_second": 12.878, |
| "step": 60 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 1.3325926065444946, |
| "learning_rate": 4.5e-05, |
| "loss": 0.5011, |
| "step": 75 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.7279411764705882, |
| "eval_combined_score": 0.7743098451012069, |
| "eval_f1": 0.8206785137318255, |
| "eval_loss": 0.5830019116401672, |
| "eval_runtime": 0.1373, |
| "eval_samples_per_second": 2972.271, |
| "eval_steps_per_second": 14.57, |
| "step": 75 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 2.0120363235473633, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.4413, |
| "step": 90 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.7058823529411765, |
| "eval_combined_score": 0.7494929006085194, |
| "eval_f1": 0.7931034482758621, |
| "eval_loss": 0.6182529330253601, |
| "eval_runtime": 0.1361, |
| "eval_samples_per_second": 2997.264, |
| "eval_steps_per_second": 14.692, |
| "step": 90 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 2.0488951206207275, |
| "learning_rate": 4.3e-05, |
| "loss": 0.3632, |
| "step": 105 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.696078431372549, |
| "eval_combined_score": 0.7443602859203882, |
| "eval_f1": 0.7926421404682275, |
| "eval_loss": 0.6797199845314026, |
| "eval_runtime": 0.1507, |
| "eval_samples_per_second": 2708.152, |
| "eval_steps_per_second": 13.275, |
| "step": 105 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 2.9908063411712646, |
| "learning_rate": 4.2e-05, |
| "loss": 0.2641, |
| "step": 120 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.7107843137254902, |
| "eval_combined_score": 0.7560655575361458, |
| "eval_f1": 0.8013468013468014, |
| "eval_loss": 0.789476215839386, |
| "eval_runtime": 0.148, |
| "eval_samples_per_second": 2757.156, |
| "eval_steps_per_second": 13.515, |
| "step": 120 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 3.745847463607788, |
| "learning_rate": 4.1e-05, |
| "loss": 0.1796, |
| "step": 135 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.6911764705882353, |
| "eval_combined_score": 0.735832137733142, |
| "eval_f1": 0.7804878048780488, |
| "eval_loss": 0.8991454839706421, |
| "eval_runtime": 0.1369, |
| "eval_samples_per_second": 2980.18, |
| "eval_steps_per_second": 14.609, |
| "step": 135 |
| }, |
| { |
| "epoch": 9.0, |
| "step": 135, |
| "total_flos": 865692760633344.0, |
| "train_loss": 0.4532281893270987, |
| "train_runtime": 26.1828, |
| "train_samples_per_second": 7004.587, |
| "train_steps_per_second": 28.645 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 750, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 50, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 5, |
| "early_stopping_threshold": 0.0 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 5 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 865692760633344.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
