| { |
| "best_global_step": 3, |
| "best_metric": 0.7056007981300354, |
| "best_model_checkpoint": "distilbert_rand_10_v2_wnli/checkpoint-3", |
| "epoch": 6.0, |
| "eval_steps": 500, |
| "global_step": 18, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.085111141204834, |
| "learning_rate": 4.9e-05, |
| "loss": 0.716, |
| "step": 3 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.4225352112676056, |
| "eval_loss": 0.7056007981300354, |
| "eval_runtime": 0.044, |
| "eval_samples_per_second": 1614.208, |
| "eval_steps_per_second": 22.735, |
| "step": 3 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.1282340288162231, |
| "learning_rate": 4.8e-05, |
| "loss": 0.7053, |
| "step": 6 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.28169014084507044, |
| "eval_loss": 0.725077211856842, |
| "eval_runtime": 0.0488, |
| "eval_samples_per_second": 1454.961, |
| "eval_steps_per_second": 20.492, |
| "step": 6 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.8648219704627991, |
| "learning_rate": 4.7e-05, |
| "loss": 0.6934, |
| "step": 9 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.4507042253521127, |
| "eval_loss": 0.728047788143158, |
| "eval_runtime": 0.0412, |
| "eval_samples_per_second": 1722.786, |
| "eval_steps_per_second": 24.265, |
| "step": 9 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 1.412021517753601, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.6861, |
| "step": 12 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.323943661971831, |
| "eval_loss": 0.7464239597320557, |
| "eval_runtime": 0.0414, |
| "eval_samples_per_second": 1716.678, |
| "eval_steps_per_second": 24.179, |
| "step": 12 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.9089189171791077, |
| "learning_rate": 4.5e-05, |
| "loss": 0.6893, |
| "step": 15 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.23943661971830985, |
| "eval_loss": 0.7720070481300354, |
| "eval_runtime": 0.0389, |
| "eval_samples_per_second": 1824.65, |
| "eval_steps_per_second": 25.699, |
| "step": 15 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 1.1620075702667236, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.6942, |
| "step": 18 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.30985915492957744, |
| "eval_loss": 0.7826804518699646, |
| "eval_runtime": 0.0418, |
| "eval_samples_per_second": 1697.121, |
| "eval_steps_per_second": 23.903, |
| "step": 18 |
| }, |
| { |
| "epoch": 6.0, |
| "step": 18, |
| "total_flos": 252350394439680.0, |
| "train_loss": 0.697368343671163, |
| "train_runtime": 11.6317, |
| "train_samples_per_second": 2729.606, |
| "train_steps_per_second": 12.896 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 150, |
| "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": 252350394439680.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
