| { | |
| "best_metric": 58.04611904264016, | |
| "best_model_checkpoint": "./models/rootflo/fauna-v0.7/checkpoint-1000", | |
| "epoch": 4.278074866310161, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0695187165775402, | |
| "grad_norm": 0.2809576392173767, | |
| "learning_rate": 3.333333333333333e-07, | |
| "loss": 0.0301, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0695187165775402, | |
| "eval_loss": 0.04982469603419304, | |
| "eval_runtime": 1269.6985, | |
| "eval_samples_per_second": 2.048, | |
| "eval_steps_per_second": 0.006, | |
| "eval_wer": 58.13926962340551, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.1390374331550803, | |
| "grad_norm": 0.27427300810813904, | |
| "learning_rate": 6.66e-07, | |
| "loss": 0.0278, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.1390374331550803, | |
| "eval_loss": 0.051491159945726395, | |
| "eval_runtime": 1268.0496, | |
| "eval_samples_per_second": 2.05, | |
| "eval_steps_per_second": 0.006, | |
| "eval_wer": 58.04611904264016, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.2085561497326203, | |
| "grad_norm": 0.3170740306377411, | |
| "learning_rate": 9.993333333333333e-07, | |
| "loss": 0.0261, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.2085561497326203, | |
| "eval_loss": 0.0537528432905674, | |
| "eval_runtime": 1273.9709, | |
| "eval_samples_per_second": 2.041, | |
| "eval_steps_per_second": 0.005, | |
| "eval_wer": 59.10879607626941, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.278074866310161, | |
| "grad_norm": 0.32307836413383484, | |
| "learning_rate": 9.506789790182363e-07, | |
| "loss": 0.0241, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.278074866310161, | |
| "eval_loss": 0.055830925703048706, | |
| "eval_runtime": 1280.7604, | |
| "eval_samples_per_second": 2.03, | |
| "eval_steps_per_second": 0.005, | |
| "eval_wer": 60.71897040092771, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 11, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.5 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 3 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.4458931356758945e+21, | |
| "train_batch_size": 72, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |