84 lines
1.9 KiB
JSON
84 lines
1.9 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.2604166666666667,
|
|
"eval_steps": 500,
|
|
"global_step": 150,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 0.034722222222222224,
|
|
"grad_norm": 2.985321283340454,
|
|
"learning_rate": 1.6379310344827587e-06,
|
|
"loss": 2.6523,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 0.06944444444444445,
|
|
"grad_norm": 2.353632926940918,
|
|
"learning_rate": 3.362068965517242e-06,
|
|
"loss": 2.4359,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 0.10416666666666667,
|
|
"grad_norm": 1.865394115447998,
|
|
"learning_rate": 5.086206896551724e-06,
|
|
"loss": 2.2545,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 0.1388888888888889,
|
|
"grad_norm": 1.8125933408737183,
|
|
"learning_rate": 6.810344827586207e-06,
|
|
"loss": 2.0557,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 0.1736111111111111,
|
|
"grad_norm": 2.3572490215301514,
|
|
"learning_rate": 8.53448275862069e-06,
|
|
"loss": 1.9334,
|
|
"step": 100
|
|
},
|
|
{
|
|
"epoch": 0.20833333333333334,
|
|
"grad_norm": 2.395277261734009,
|
|
"learning_rate": 9.999793100349294e-06,
|
|
"loss": 1.9212,
|
|
"step": 120
|
|
},
|
|
{
|
|
"epoch": 0.24305555555555555,
|
|
"grad_norm": 2.714841604232788,
|
|
"learning_rate": 9.987843743451796e-06,
|
|
"loss": 1.7889,
|
|
"step": 140
|
|
}
|
|
],
|
|
"logging_steps": 20,
|
|
"max_steps": 1152,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 2,
|
|
"save_steps": 50,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 3.049872287249203e+16,
|
|
"train_batch_size": 4,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|