|
{ |
|
"best_metric": 0.645350456237793, |
|
"best_model_checkpoint": "miner_id_24/checkpoint-100", |
|
"epoch": 0.007984908522891734, |
|
"eval_steps": 50, |
|
"global_step": 100, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 7.984908522891735e-05, |
|
"eval_loss": 2.0086100101470947, |
|
"eval_runtime": 702.0595, |
|
"eval_samples_per_second": 15.023, |
|
"eval_steps_per_second": 7.512, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.003193963409156694, |
|
"grad_norm": 60.47791290283203, |
|
"learning_rate": 5e-05, |
|
"loss": 8.1173, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.003992454261445867, |
|
"eval_loss": 0.7402405142784119, |
|
"eval_runtime": 703.0805, |
|
"eval_samples_per_second": 15.001, |
|
"eval_steps_per_second": 7.501, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.006387926818313388, |
|
"grad_norm": 37.602317810058594, |
|
"learning_rate": 0.0001, |
|
"loss": 5.2458, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.007984908522891734, |
|
"eval_loss": 0.645350456237793, |
|
"eval_runtime": 703.8766, |
|
"eval_samples_per_second": 14.984, |
|
"eval_steps_per_second": 7.493, |
|
"step": 100 |
|
} |
|
], |
|
"logging_steps": 40, |
|
"max_steps": 400, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 50, |
|
"stateful_callbacks": { |
|
"EarlyStoppingCallback": { |
|
"args": { |
|
"early_stopping_patience": 3, |
|
"early_stopping_threshold": 0.0 |
|
}, |
|
"attributes": { |
|
"early_stopping_patience_counter": 0 |
|
} |
|
}, |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 3.5406836655980544e+16, |
|
"train_batch_size": 2, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|