|
{ |
|
"best_metric": NaN, |
|
"best_model_checkpoint": "miner_id_24/checkpoint-10", |
|
"epoch": 0.0014657920773938217, |
|
"eval_steps": 5, |
|
"global_step": 10, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.00014657920773938218, |
|
"eval_loss": NaN, |
|
"eval_runtime": 684.4397, |
|
"eval_samples_per_second": 4.198, |
|
"eval_steps_per_second": 2.1, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.0004397376232181465, |
|
"grad_norm": NaN, |
|
"learning_rate": 0.00012, |
|
"loss": 0.0, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 0.0007328960386969108, |
|
"eval_loss": NaN, |
|
"eval_runtime": 682.5225, |
|
"eval_samples_per_second": 4.209, |
|
"eval_steps_per_second": 2.105, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.000879475246436293, |
|
"grad_norm": NaN, |
|
"learning_rate": 0.0001992114701314478, |
|
"loss": 0.0, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 0.0013192128696544396, |
|
"grad_norm": NaN, |
|
"learning_rate": 0.00018763066800438636, |
|
"loss": 0.0, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 0.0014657920773938217, |
|
"eval_loss": NaN, |
|
"eval_runtime": 682.9124, |
|
"eval_samples_per_second": 4.207, |
|
"eval_steps_per_second": 2.104, |
|
"step": 10 |
|
} |
|
], |
|
"logging_steps": 3, |
|
"max_steps": 30, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 10, |
|
"stateful_callbacks": { |
|
"EarlyStoppingCallback": { |
|
"args": { |
|
"early_stopping_patience": 1, |
|
"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": 4079510832021504.0, |
|
"train_batch_size": 2, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|