| { |
| "best_global_step": 820, |
| "best_metric": 0.6217690110206604, |
| "best_model_checkpoint": "distilbert_km_5_v1_qnli/checkpoint-820", |
| "epoch": 7.0, |
| "eval_steps": 500, |
| "global_step": 2870, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.8706941604614258, |
| "learning_rate": 4.9e-05, |
| "loss": 0.6628, |
| "step": 410 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6428702178290316, |
| "eval_loss": 0.6315329670906067, |
| "eval_runtime": 2.2428, |
| "eval_samples_per_second": 2435.799, |
| "eval_steps_per_second": 9.809, |
| "step": 410 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 2.4346811771392822, |
| "learning_rate": 4.8e-05, |
| "loss": 0.612, |
| "step": 820 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.652022698151199, |
| "eval_loss": 0.6217690110206604, |
| "eval_runtime": 2.206, |
| "eval_samples_per_second": 2476.483, |
| "eval_steps_per_second": 9.973, |
| "step": 820 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 3.585360050201416, |
| "learning_rate": 4.7e-05, |
| "loss": 0.527, |
| "step": 1230 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.643053267435475, |
| "eval_loss": 0.6505500078201294, |
| "eval_runtime": 2.2104, |
| "eval_samples_per_second": 2471.551, |
| "eval_steps_per_second": 9.953, |
| "step": 1230 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 6.053065299987793, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.4147, |
| "step": 1640 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.6500091524803222, |
| "eval_loss": 0.7125176191329956, |
| "eval_runtime": 2.2137, |
| "eval_samples_per_second": 2467.811, |
| "eval_steps_per_second": 9.938, |
| "step": 1640 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 8.080596923828125, |
| "learning_rate": 4.5e-05, |
| "loss": 0.3011, |
| "step": 2050 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6512904997254256, |
| "eval_loss": 0.8449434041976929, |
| "eval_runtime": 2.2192, |
| "eval_samples_per_second": 2461.687, |
| "eval_steps_per_second": 9.913, |
| "step": 2050 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 7.3134331703186035, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.2112, |
| "step": 2460 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.6454329123192385, |
| "eval_loss": 1.0667665004730225, |
| "eval_runtime": 2.2407, |
| "eval_samples_per_second": 2438.102, |
| "eval_steps_per_second": 9.818, |
| "step": 2460 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 10.40584945678711, |
| "learning_rate": 4.3e-05, |
| "loss": 0.1582, |
| "step": 2870 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.6353651839648545, |
| "eval_loss": 1.2523982524871826, |
| "eval_runtime": 2.2217, |
| "eval_samples_per_second": 2458.979, |
| "eval_steps_per_second": 9.903, |
| "step": 2870 |
| }, |
| { |
| "epoch": 7.0, |
| "step": 2870, |
| "total_flos": 4.856261458098893e+16, |
| "train_loss": 0.4124086280317672, |
| "train_runtime": 691.1366, |
| "train_samples_per_second": 7577.59, |
| "train_steps_per_second": 29.661 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 20500, |
| "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": 4.856261458098893e+16, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
