107 lines
2.6 KiB
JSON
107 lines
2.6 KiB
JSON
|
{
|
||
|
|
"best_global_step": 2536,
|
||
|
|
"best_metric": 76.57829761930071,
|
||
|
|
"best_model_checkpoint": "mbert-loraxs-qa-vanilla/run-1/checkpoint-2536",
|
||
|
|
"epoch": 4.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 2536,
|
||
|
|
"is_hyper_param_search": true,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"grad_norm": 15.868330955505371,
|
||
|
|
"learning_rate": 5.191675659155908e-05,
|
||
|
|
"loss": 2.1192049333349763,
|
||
|
|
"step": 634
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_exact_match": 60.08,
|
||
|
|
"eval_f1": 74.33783203646502,
|
||
|
|
"eval_loss": 1.2725635766983032,
|
||
|
|
"eval_runtime": 5.8013,
|
||
|
|
"eval_samples_per_second": 218.055,
|
||
|
|
"eval_steps_per_second": 13.79,
|
||
|
|
"step": 634
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.0,
|
||
|
|
"grad_norm": 30.633678436279297,
|
||
|
|
"learning_rate": 3.46202649052488e-05,
|
||
|
|
"loss": 0.9956009440617607,
|
||
|
|
"step": 1268
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.0,
|
||
|
|
"eval_exact_match": 62.48,
|
||
|
|
"eval_f1": 76.03924855668853,
|
||
|
|
"eval_loss": 1.3048548698425293,
|
||
|
|
"eval_runtime": 6.0639,
|
||
|
|
"eval_samples_per_second": 208.613,
|
||
|
|
"eval_steps_per_second": 13.193,
|
||
|
|
"step": 1268
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.0,
|
||
|
|
"grad_norm": 13.438405990600586,
|
||
|
|
"learning_rate": 1.7323773218938526e-05,
|
||
|
|
"loss": 0.49370124889097006,
|
||
|
|
"step": 1902
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.0,
|
||
|
|
"eval_exact_match": 62.8,
|
||
|
|
"eval_f1": 76.44229300202514,
|
||
|
|
"eval_loss": 1.4875582456588745,
|
||
|
|
"eval_runtime": 5.797,
|
||
|
|
"eval_samples_per_second": 218.217,
|
||
|
|
"eval_steps_per_second": 13.8,
|
||
|
|
"step": 1902
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 4.0,
|
||
|
|
"grad_norm": 14.641495704650879,
|
||
|
|
"learning_rate": 2.728153262824965e-08,
|
||
|
|
"loss": 0.23204485775920494,
|
||
|
|
"step": 2536
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 4.0,
|
||
|
|
"eval_exact_match": 62.96,
|
||
|
|
"eval_f1": 76.57829761930071,
|
||
|
|
"eval_loss": 1.8072655200958252,
|
||
|
|
"eval_runtime": 5.8611,
|
||
|
|
"eval_samples_per_second": 215.829,
|
||
|
|
"eval_steps_per_second": 13.649,
|
||
|
|
"step": 2536
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 500,
|
||
|
|
"max_steps": 2536,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 4,
|
||
|
|
"save_steps": 50,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": true
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 7950999010719744.0,
|
||
|
|
"train_batch_size": 16,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": {
|
||
|
|
"learning_rate": 6.22564574576657e-05,
|
||
|
|
"num_train_epochs": 4,
|
||
|
|
"weight_decay": 0.033952923108431954
|
||
|
|
}
|
||
|
|
}
|