| { |
| "best_global_step": 30, |
| "best_metric": 0.5995076298713684, |
| "best_model_checkpoint": "bert_base_km_10_v2_mrpc/checkpoint-30", |
| "epoch": 7.0, |
| "eval_steps": 500, |
| "global_step": 105, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.238810420036316, |
| "learning_rate": 4.9e-05, |
| "loss": 0.6267, |
| "step": 15 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6764705882352942, |
| "eval_combined_score": 0.7341343477454073, |
| "eval_f1": 0.7917981072555205, |
| "eval_loss": 0.610476016998291, |
| "eval_runtime": 0.2693, |
| "eval_samples_per_second": 1515.201, |
| "eval_steps_per_second": 7.427, |
| "step": 15 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.8361072540283203, |
| "learning_rate": 4.8e-05, |
| "loss": 0.5585, |
| "step": 30 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.6985294117647058, |
| "eval_combined_score": 0.7530205744269538, |
| "eval_f1": 0.8075117370892019, |
| "eval_loss": 0.5995076298713684, |
| "eval_runtime": 0.2796, |
| "eval_samples_per_second": 1459.167, |
| "eval_steps_per_second": 7.153, |
| "step": 30 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.5062758922576904, |
| "learning_rate": 4.7e-05, |
| "loss": 0.4819, |
| "step": 45 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.6813725490196079, |
| "eval_combined_score": 0.7371830897964282, |
| "eval_f1": 0.7929936305732485, |
| "eval_loss": 0.627225399017334, |
| "eval_runtime": 0.2697, |
| "eval_samples_per_second": 1513.047, |
| "eval_steps_per_second": 7.417, |
| "step": 45 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 3.018381357192993, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.3844, |
| "step": 60 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.6862745098039216, |
| "eval_combined_score": 0.7346626786307744, |
| "eval_f1": 0.7830508474576271, |
| "eval_loss": 0.6860804557800293, |
| "eval_runtime": 0.2718, |
| "eval_samples_per_second": 1500.886, |
| "eval_steps_per_second": 7.357, |
| "step": 60 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 3.4878857135772705, |
| "learning_rate": 4.5e-05, |
| "loss": 0.2477, |
| "step": 75 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6764705882352942, |
| "eval_combined_score": 0.7207975716977182, |
| "eval_f1": 0.7651245551601423, |
| "eval_loss": 0.81784588098526, |
| "eval_runtime": 0.2717, |
| "eval_samples_per_second": 1501.599, |
| "eval_steps_per_second": 7.361, |
| "step": 75 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 2.1889967918395996, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.1321, |
| "step": 90 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.678921568627451, |
| "eval_combined_score": 0.7247497510387693, |
| "eval_f1": 0.7705779334500875, |
| "eval_loss": 1.0816551446914673, |
| "eval_runtime": 0.2808, |
| "eval_samples_per_second": 1452.733, |
| "eval_steps_per_second": 7.121, |
| "step": 90 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 2.3499011993408203, |
| "learning_rate": 4.3e-05, |
| "loss": 0.0645, |
| "step": 105 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.6862745098039216, |
| "eval_combined_score": 0.7288515406162466, |
| "eval_f1": 0.7714285714285715, |
| "eval_loss": 1.293560266494751, |
| "eval_runtime": 0.2908, |
| "eval_samples_per_second": 1402.882, |
| "eval_steps_per_second": 6.877, |
| "step": 105 |
| }, |
| { |
| "epoch": 7.0, |
| "step": 105, |
| "total_flos": 3377819728711680.0, |
| "train_loss": 0.3565297211919512, |
| "train_runtime": 57.4934, |
| "train_samples_per_second": 3189.931, |
| "train_steps_per_second": 13.045 |
| } |
| ], |
| "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": 3377819728711680.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
