76 lines
1.9 KiB
JSON
76 lines
1.9 KiB
JSON
{
|
|
"best_global_step": 1268,
|
|
"best_metric": 73.18016605585976,
|
|
"best_model_checkpoint": "mbert-loraxs-qa-vanilla/run-2/checkpoint-1268",
|
|
"epoch": 2.0,
|
|
"eval_steps": 500,
|
|
"global_step": 1268,
|
|
"is_hyper_param_search": true,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 1.0,
|
|
"grad_norm": 7.677847385406494,
|
|
"learning_rate": 0.0002951562058189359,
|
|
"loss": 3.075869587317626,
|
|
"step": 634
|
|
},
|
|
{
|
|
"epoch": 1.0,
|
|
"eval_exact_match": 56.24,
|
|
"eval_f1": 70.50074993342362,
|
|
"eval_loss": 1.4242188930511475,
|
|
"eval_runtime": 7.8142,
|
|
"eval_samples_per_second": 161.884,
|
|
"eval_steps_per_second": 10.238,
|
|
"step": 634
|
|
},
|
|
{
|
|
"epoch": 2.0,
|
|
"grad_norm": 5.153384208679199,
|
|
"learning_rate": 0.00023614358062524137,
|
|
"loss": 1.4205032493037757,
|
|
"step": 1268
|
|
},
|
|
{
|
|
"epoch": 2.0,
|
|
"eval_exact_match": 59.52,
|
|
"eval_f1": 73.18016605585976,
|
|
"eval_loss": 1.3157330751419067,
|
|
"eval_runtime": 7.7666,
|
|
"eval_samples_per_second": 162.876,
|
|
"eval_steps_per_second": 10.3,
|
|
"step": 1268
|
|
}
|
|
],
|
|
"logging_steps": 500,
|
|
"max_steps": 3804,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 6,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 6180991462324224.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": {
|
|
"learning_rate": 0.0003186123281356725,
|
|
"lora_alpha": 64,
|
|
"lora_dropout": 0.0985087648472941,
|
|
"lora_r": 256,
|
|
"num_train_epochs": 6,
|
|
"weight_decay": 0.08602299412879012
|
|
}
|
|
}
|