93 lines
2.3 KiB
JSON
93 lines
2.3 KiB
JSON
|
{
|
||
|
|
"best_global_step": 1902,
|
||
|
|
"best_metric": 74.55959025020111,
|
||
|
|
"best_model_checkpoint": "mbert-loraxs-qa-vanilla/run-7/checkpoint-1902",
|
||
|
|
"epoch": 3.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 1902,
|
||
|
|
"is_hyper_param_search": true,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"grad_norm": 56.062278747558594,
|
||
|
|
"learning_rate": 5.730149137854825e-05,
|
||
|
|
"loss": 3.2366831686218456,
|
||
|
|
"step": 634
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_exact_match": 52.72,
|
||
|
|
"eval_f1": 66.45284613864096,
|
||
|
|
"eval_loss": 1.5741198062896729,
|
||
|
|
"eval_runtime": 7.702,
|
||
|
|
"eval_samples_per_second": 164.243,
|
||
|
|
"eval_steps_per_second": 10.387,
|
||
|
|
"step": 634
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.0,
|
||
|
|
"grad_norm": 67.15625,
|
||
|
|
"learning_rate": 3.821103129762361e-05,
|
||
|
|
"loss": 1.582933780898807,
|
||
|
|
"step": 1268
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.0,
|
||
|
|
"eval_exact_match": 57.28,
|
||
|
|
"eval_f1": 71.03664562956527,
|
||
|
|
"eval_loss": 1.4219049215316772,
|
||
|
|
"eval_runtime": 7.7459,
|
||
|
|
"eval_samples_per_second": 163.312,
|
||
|
|
"eval_steps_per_second": 10.328,
|
||
|
|
"step": 1268
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.0,
|
||
|
|
"grad_norm": 40.657630920410156,
|
||
|
|
"learning_rate": 1.9120571216698972e-05,
|
||
|
|
"loss": 1.2915771676914924,
|
||
|
|
"step": 1902
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.0,
|
||
|
|
"eval_exact_match": 60.64,
|
||
|
|
"eval_f1": 74.55959025020111,
|
||
|
|
"eval_loss": 1.3220536708831787,
|
||
|
|
"eval_runtime": 7.7105,
|
||
|
|
"eval_samples_per_second": 164.063,
|
||
|
|
"eval_steps_per_second": 10.375,
|
||
|
|
"step": 1902
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 500,
|
||
|
|
"max_steps": 2536,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 4,
|
||
|
|
"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": 7534718886417408.0,
|
||
|
|
"train_batch_size": 16,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": {
|
||
|
|
"learning_rate": 6.871361183701897e-05,
|
||
|
|
"lora_alpha": 512,
|
||
|
|
"lora_dropout": 0.2970595529882041,
|
||
|
|
"lora_r": 128,
|
||
|
|
"num_train_epochs": 4,
|
||
|
|
"weight_decay": 0.0640462328905566
|
||
|
|
}
|
||
|
|
}
|