102 lines
2.4 KiB
JSON
102 lines
2.4 KiB
JSON
|
{
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 3.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 21,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.14285714285714285,
|
||
|
|
"grad_norm": 99.75653578103409,
|
||
|
|
"learning_rate": 6.666666666666667e-06,
|
||
|
|
"loss": 1.7828,
|
||
|
|
"step": 1
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7142857142857143,
|
||
|
|
"grad_norm": 421.35114809358015,
|
||
|
|
"learning_rate": 1.9396926207859085e-05,
|
||
|
|
"loss": 1.8602,
|
||
|
|
"step": 5
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_loss": 1.3279948234558105,
|
||
|
|
"eval_runtime": 2.2021,
|
||
|
|
"eval_samples_per_second": 39.054,
|
||
|
|
"eval_steps_per_second": 0.908,
|
||
|
|
"step": 7
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.4285714285714286,
|
||
|
|
"grad_norm": 60.4301410351058,
|
||
|
|
"learning_rate": 1.342020143325669e-05,
|
||
|
|
"loss": 2.9102,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.0,
|
||
|
|
"eval_loss": 0.6256123781204224,
|
||
|
|
"eval_runtime": 2.0607,
|
||
|
|
"eval_samples_per_second": 41.733,
|
||
|
|
"eval_steps_per_second": 0.971,
|
||
|
|
"step": 14
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.142857142857143,
|
||
|
|
"grad_norm": 9.507888146594869,
|
||
|
|
"learning_rate": 5.000000000000003e-06,
|
||
|
|
"loss": 0.686,
|
||
|
|
"step": 15
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.857142857142857,
|
||
|
|
"grad_norm": 1.2620367321960468,
|
||
|
|
"learning_rate": 1.519224698779198e-07,
|
||
|
|
"loss": 0.5329,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.0,
|
||
|
|
"eval_loss": 0.5474353432655334,
|
||
|
|
"eval_runtime": 2.0545,
|
||
|
|
"eval_samples_per_second": 41.86,
|
||
|
|
"eval_steps_per_second": 0.973,
|
||
|
|
"step": 21
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.0,
|
||
|
|
"step": 21,
|
||
|
|
"total_flos": 8793945538560.0,
|
||
|
|
"train_loss": 1.4474536804925828,
|
||
|
|
"train_runtime": 170.8306,
|
||
|
|
"train_samples_per_second": 13.821,
|
||
|
|
"train_steps_per_second": 0.123
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 5,
|
||
|
|
"max_steps": 21,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 3,
|
||
|
|
"save_steps": 100,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": false,
|
||
|
|
"should_training_stop": false
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 8793945538560.0,
|
||
|
|
"train_batch_size": 16,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|