| { |
| "best_global_step": 820, |
| "best_metric": 0.6339648962020874, |
| "best_model_checkpoint": "tiny_bert_rand_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": 1.6268937587738037, |
| "learning_rate": 4.9e-05, |
| "loss": 0.6605, |
| "step": 410 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6243822075782537, |
| "eval_loss": 0.641603946685791, |
| "eval_runtime": 1.6314, |
| "eval_samples_per_second": 3348.738, |
| "eval_steps_per_second": 13.486, |
| "step": 410 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.7828949689865112, |
| "learning_rate": 4.8e-05, |
| "loss": 0.6264, |
| "step": 820 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.6393922753066081, |
| "eval_loss": 0.6339648962020874, |
| "eval_runtime": 1.606, |
| "eval_samples_per_second": 3401.523, |
| "eval_steps_per_second": 13.698, |
| "step": 820 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 3.362802267074585, |
| "learning_rate": 4.7e-05, |
| "loss": 0.5715, |
| "step": 1230 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.6353651839648545, |
| "eval_loss": 0.6345011591911316, |
| "eval_runtime": 1.607, |
| "eval_samples_per_second": 3399.573, |
| "eval_steps_per_second": 13.69, |
| "step": 1230 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 4.2415242195129395, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.4897, |
| "step": 1640 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.6401244737323815, |
| "eval_loss": 0.6881918907165527, |
| "eval_runtime": 1.5912, |
| "eval_samples_per_second": 3433.251, |
| "eval_steps_per_second": 13.826, |
| "step": 1640 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 4.094640731811523, |
| "learning_rate": 4.5e-05, |
| "loss": 0.4121, |
| "step": 2050 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6362804319970712, |
| "eval_loss": 0.7537786960601807, |
| "eval_runtime": 1.5957, |
| "eval_samples_per_second": 3423.643, |
| "eval_steps_per_second": 13.787, |
| "step": 2050 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 6.535858631134033, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.3443, |
| "step": 2460 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.6386600768808347, |
| "eval_loss": 0.8461039662361145, |
| "eval_runtime": 1.6159, |
| "eval_samples_per_second": 3380.718, |
| "eval_steps_per_second": 13.614, |
| "step": 2460 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 7.503296375274658, |
| "learning_rate": 4.3e-05, |
| "loss": 0.2845, |
| "step": 2870 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.6285923485264506, |
| "eval_loss": 1.0166436433792114, |
| "eval_runtime": 1.6294, |
| "eval_samples_per_second": 3352.801, |
| "eval_steps_per_second": 13.502, |
| "step": 2870 |
| }, |
| { |
| "epoch": 7.0, |
| "step": 2870, |
| "total_flos": 1.922715369529651e+16, |
| "train_loss": 0.48413124217388936, |
| "train_runtime": 404.2087, |
| "train_samples_per_second": 12956.55, |
| "train_steps_per_second": 50.716 |
| } |
| ], |
| "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": 1.922715369529651e+16, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
