127 lines
3.2 KiB
JSON
127 lines
3.2 KiB
JSON
|
{
|
||
|
|
"best_global_step": 3170,
|
||
|
|
"best_metric": 76.77441671821548,
|
||
|
|
"best_model_checkpoint": "mbert-loraxs-qa-vanilla/run-12/checkpoint-3170",
|
||
|
|
"epoch": 5.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 3170,
|
||
|
|
"is_hyper_param_search": true,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"grad_norm": 4.95363712310791,
|
||
|
|
"learning_rate": 0.00043918755006463925,
|
||
|
|
"loss": 2.7746150741817823,
|
||
|
|
"step": 634
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_exact_match": 57.12,
|
||
|
|
"eval_f1": 71.22068270961232,
|
||
|
|
"eval_loss": 1.3444347381591797,
|
||
|
|
"eval_runtime": 7.7895,
|
||
|
|
"eval_samples_per_second": 162.397,
|
||
|
|
"eval_steps_per_second": 10.27,
|
||
|
|
"step": 634
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.0,
|
||
|
|
"grad_norm": 4.687035083770752,
|
||
|
|
"learning_rate": 0.00032943394078557684,
|
||
|
|
"loss": 1.3028373838599172,
|
||
|
|
"step": 1268
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.0,
|
||
|
|
"eval_exact_match": 59.44,
|
||
|
|
"eval_f1": 74.2554549397151,
|
||
|
|
"eval_loss": 1.300154447555542,
|
||
|
|
"eval_runtime": 7.784,
|
||
|
|
"eval_samples_per_second": 162.512,
|
||
|
|
"eval_steps_per_second": 10.277,
|
||
|
|
"step": 1268
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.0,
|
||
|
|
"grad_norm": 4.485013008117676,
|
||
|
|
"learning_rate": 0.00021968033150651447,
|
||
|
|
"loss": 0.9939027419225651,
|
||
|
|
"step": 1902
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.0,
|
||
|
|
"eval_exact_match": 62.32,
|
||
|
|
"eval_f1": 76.46467374081998,
|
||
|
|
"eval_loss": 1.2457834482192993,
|
||
|
|
"eval_runtime": 7.8041,
|
||
|
|
"eval_samples_per_second": 162.095,
|
||
|
|
"eval_steps_per_second": 10.251,
|
||
|
|
"step": 1902
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 4.0,
|
||
|
|
"grad_norm": 5.448144912719727,
|
||
|
|
"learning_rate": 0.00010992672222745207,
|
||
|
|
"loss": 0.7817085819665566,
|
||
|
|
"step": 2536
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 4.0,
|
||
|
|
"eval_exact_match": 62.56,
|
||
|
|
"eval_f1": 76.3891847215123,
|
||
|
|
"eval_loss": 1.305019736289978,
|
||
|
|
"eval_runtime": 7.8165,
|
||
|
|
"eval_samples_per_second": 161.837,
|
||
|
|
"eval_steps_per_second": 10.235,
|
||
|
|
"step": 2536
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 5.0,
|
||
|
|
"grad_norm": 2.5814151763916016,
|
||
|
|
"learning_rate": 1.7311294838968833e-07,
|
||
|
|
"loss": 0.6479539841131458,
|
||
|
|
"step": 3170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 5.0,
|
||
|
|
"eval_exact_match": 62.32,
|
||
|
|
"eval_f1": 76.77441671821548,
|
||
|
|
"eval_loss": 1.3857749700546265,
|
||
|
|
"eval_runtime": 7.7727,
|
||
|
|
"eval_samples_per_second": 162.749,
|
||
|
|
"eval_steps_per_second": 10.292,
|
||
|
|
"step": 3170
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 500,
|
||
|
|
"max_steps": 3170,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 5,
|
||
|
|
"save_steps": 500,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": true
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 1.545247865581056e+16,
|
||
|
|
"train_batch_size": 16,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": {
|
||
|
|
"learning_rate": 0.0004938912417557808,
|
||
|
|
"lora_alpha": 64,
|
||
|
|
"lora_dropout": 0.08091352388249862,
|
||
|
|
"lora_r": 256,
|
||
|
|
"num_train_epochs": 5,
|
||
|
|
"weight_decay": 0.04137309253112481
|
||
|
|
}
|
||
|
|
}
|