| { |
| "best_global_step": 3, |
| "best_metric": 0.7345951795578003, |
| "best_model_checkpoint": "distilbert_km_50_v1_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": 1.2527297735214233, |
| "learning_rate": 4.9e-05, |
| "loss": 0.728, |
| "step": 3 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.38028169014084506, |
| "eval_loss": 0.7345951795578003, |
| "eval_runtime": 0.0392, |
| "eval_samples_per_second": 1809.219, |
| "eval_steps_per_second": 25.482, |
| "step": 3 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.0723785161972046, |
| "learning_rate": 4.8e-05, |
| "loss": 0.6997, |
| "step": 6 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.30985915492957744, |
| "eval_loss": 0.7613610029220581, |
| "eval_runtime": 0.0406, |
| "eval_samples_per_second": 1749.68, |
| "eval_steps_per_second": 24.643, |
| "step": 6 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.0101525783538818, |
| "learning_rate": 4.7e-05, |
| "loss": 0.6966, |
| "step": 9 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.323943661971831, |
| "eval_loss": 0.7619937658309937, |
| "eval_runtime": 0.0477, |
| "eval_samples_per_second": 1489.633, |
| "eval_steps_per_second": 20.981, |
| "step": 9 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 1.5457305908203125, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.6881, |
| "step": 12 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.36619718309859156, |
| "eval_loss": 0.7787742018699646, |
| "eval_runtime": 0.0451, |
| "eval_samples_per_second": 1574.339, |
| "eval_steps_per_second": 22.174, |
| "step": 12 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 1.316327452659607, |
| "learning_rate": 4.5e-05, |
| "loss": 0.6847, |
| "step": 15 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.23943661971830985, |
| "eval_loss": 0.8101891279220581, |
| "eval_runtime": 0.0433, |
| "eval_samples_per_second": 1638.382, |
| "eval_steps_per_second": 23.076, |
| "step": 15 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 1.3633482456207275, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.6805, |
| "step": 18 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.2112676056338028, |
| "eval_loss": 0.843089759349823, |
| "eval_runtime": 0.0446, |
| "eval_samples_per_second": 1591.196, |
| "eval_steps_per_second": 22.411, |
| "step": 18 |
| }, |
| { |
| "epoch": 6.0, |
| "step": 18, |
| "total_flos": 252350394439680.0, |
| "train_loss": 0.696275896496243, |
| "train_runtime": 11.7606, |
| "train_samples_per_second": 2699.684, |
| "train_steps_per_second": 12.754 |
| } |
| ], |
| "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 |
| } |
|
|
