| { |
| "best_global_step": 6136, |
| "best_metric": 0.790632963180542, |
| "best_model_checkpoint": "distilbert_km_10_v2_mnli/checkpoint-6136", |
| "epoch": 9.0, |
| "eval_steps": 500, |
| "global_step": 13806, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.0762512683868408, |
| "learning_rate": 4.9e-05, |
| "loss": 0.9871, |
| "step": 1534 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.5689251146204789, |
| "eval_loss": 0.9118844866752625, |
| "eval_runtime": 4.3355, |
| "eval_samples_per_second": 2263.846, |
| "eval_steps_per_second": 8.995, |
| "step": 1534 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.6051461696624756, |
| "learning_rate": 4.8e-05, |
| "loss": 0.8774, |
| "step": 3068 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.6142638818135507, |
| "eval_loss": 0.8434556126594543, |
| "eval_runtime": 3.933, |
| "eval_samples_per_second": 2495.575, |
| "eval_steps_per_second": 9.916, |
| "step": 3068 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.8933354616165161, |
| "learning_rate": 4.7e-05, |
| "loss": 0.7905, |
| "step": 4602 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.6473764645950076, |
| "eval_loss": 0.8013252019882202, |
| "eval_runtime": 4.0002, |
| "eval_samples_per_second": 2453.623, |
| "eval_steps_per_second": 9.749, |
| "step": 4602 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 2.1612749099731445, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.71, |
| "step": 6136 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.661334691798268, |
| "eval_loss": 0.790632963180542, |
| "eval_runtime": 4.0008, |
| "eval_samples_per_second": 2453.246, |
| "eval_steps_per_second": 9.748, |
| "step": 6136 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 2.4366869926452637, |
| "learning_rate": 4.5e-05, |
| "loss": 0.6392, |
| "step": 7670 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6657157412124299, |
| "eval_loss": 0.8011903762817383, |
| "eval_runtime": 3.9757, |
| "eval_samples_per_second": 2468.768, |
| "eval_steps_per_second": 9.81, |
| "step": 7670 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 2.650559186935425, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.5668, |
| "step": 9204 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.6630667345899134, |
| "eval_loss": 0.8617571592330933, |
| "eval_runtime": 3.9973, |
| "eval_samples_per_second": 2455.417, |
| "eval_steps_per_second": 9.757, |
| "step": 9204 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 2.723828077316284, |
| "learning_rate": 4.3e-05, |
| "loss": 0.4973, |
| "step": 10738 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.6631686194600102, |
| "eval_loss": 0.8917017579078674, |
| "eval_runtime": 3.926, |
| "eval_samples_per_second": 2500.014, |
| "eval_steps_per_second": 9.934, |
| "step": 10738 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 2.998859405517578, |
| "learning_rate": 4.2e-05, |
| "loss": 0.4303, |
| "step": 12272 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.6532857870606215, |
| "eval_loss": 1.0054373741149902, |
| "eval_runtime": 3.9713, |
| "eval_samples_per_second": 2471.477, |
| "eval_steps_per_second": 9.82, |
| "step": 12272 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 3.5228171348571777, |
| "learning_rate": 4.1e-05, |
| "loss": 0.3693, |
| "step": 13806 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.6584819154355578, |
| "eval_loss": 1.1392618417739868, |
| "eval_runtime": 4.0353, |
| "eval_samples_per_second": 2432.296, |
| "eval_steps_per_second": 9.665, |
| "step": 13806 |
| }, |
| { |
| "epoch": 9.0, |
| "step": 13806, |
| "total_flos": 2.340951304214108e+17, |
| "train_loss": 0.6519948513667065, |
| "train_runtime": 3244.5973, |
| "train_samples_per_second": 6051.629, |
| "train_steps_per_second": 23.639 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 76700, |
| "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": 2.340951304214108e+17, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
