| { |
| "best_metric": 39.990222439501345, |
| "best_model_checkpoint": "./base/checkpoint-500", |
| "epoch": 1.8903591682419658, |
| "eval_steps": 500, |
| "global_step": 2000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.000945179584120983, |
| "grad_norm": 7431396.5, |
| "learning_rate": 5.000000000000001e-07, |
| "loss": 1.9363, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.4725897920604915, |
| "grad_norm": 54115.1796875, |
| "learning_rate": 1.923076923076923e-05, |
| "loss": 0.2641, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.4725897920604915, |
| "eval_loss": 0.056086380034685135, |
| "eval_runtime": 3242.1523, |
| "eval_samples_per_second": 1.015, |
| "eval_steps_per_second": 0.127, |
| "eval_wer": 39.990222439501345, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.945179584120983, |
| "grad_norm": 42365.73828125, |
| "learning_rate": 1.282051282051282e-05, |
| "loss": 0.0487, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.945179584120983, |
| "eval_loss": 0.041050925850868225, |
| "eval_runtime": 3191.4517, |
| "eval_samples_per_second": 1.031, |
| "eval_steps_per_second": 0.129, |
| "eval_wer": 31.500667506534043, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.4177693761814745, |
| "grad_norm": 46034.90625, |
| "learning_rate": 6.41025641025641e-06, |
| "loss": 0.0331, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.4177693761814745, |
| "eval_loss": 0.036172620952129364, |
| "eval_runtime": 3173.1494, |
| "eval_samples_per_second": 1.037, |
| "eval_steps_per_second": 0.13, |
| "eval_wer": 28.456461651279547, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.8903591682419658, |
| "grad_norm": 35570.5625, |
| "learning_rate": 0.0, |
| "loss": 0.0302, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.8903591682419658, |
| "eval_loss": 0.03396076709032059, |
| "eval_runtime": 3170.9478, |
| "eval_samples_per_second": 1.038, |
| "eval_steps_per_second": 0.13, |
| "eval_wer": 27.166575785495368, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 2000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.63125174501376e+18, |
| "train_batch_size": 28, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
