84 lines
2.1 KiB
JSON
84 lines
2.1 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.0,
|
|
"eval_steps": 400,
|
|
"global_step": 87,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 1.872968821413815,
|
|
"epoch": 0.23238925199709515,
|
|
"grad_norm": 0.6044484972953796,
|
|
"learning_rate": 2.7393581614739924e-05,
|
|
"loss": 2.7303,
|
|
"mean_token_accuracy": 0.5062888355925679,
|
|
"num_tokens": 625977.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 2.1671952702105046,
|
|
"epoch": 0.4647785039941903,
|
|
"grad_norm": 0.477797269821167,
|
|
"learning_rate": 1.8337814009344716e-05,
|
|
"loss": 2.3616,
|
|
"mean_token_accuracy": 0.5430196014232933,
|
|
"num_tokens": 1260638.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 2.2003143060952426,
|
|
"epoch": 0.6971677559912854,
|
|
"grad_norm": 0.4428477883338928,
|
|
"learning_rate": 7.500000000000004e-06,
|
|
"loss": 2.2211,
|
|
"mean_token_accuracy": 0.5593088129535317,
|
|
"num_tokens": 1885459.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 2.1266778625547884,
|
|
"epoch": 0.9295570079883806,
|
|
"grad_norm": 0.3918190598487854,
|
|
"learning_rate": 6.664079132078881e-07,
|
|
"loss": 2.1177,
|
|
"mean_token_accuracy": 0.5764182968065142,
|
|
"num_tokens": 2516412.0,
|
|
"step": 80
|
|
}
|
|
],
|
|
"logging_steps": 20,
|
|
"max_steps": 87,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 400,
|
|
"stateful_callbacks": {
|
|
"EarlyStoppingCallback": {
|
|
"args": {
|
|
"early_stopping_patience": 2,
|
|
"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": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 6.151106844226368e+16,
|
|
"train_batch_size": 2,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|