NT1-500M-human_ref-finetuned-H4ac / trainer_state.json
Chenghao-Qiu's picture
Upload folder using huggingface_hub
365fbb0 verified
raw
history blame
2.58 kB
{
"best_metric": 0.6257979869842529,
"best_model_checkpoint": "output_pipe/H4ac/origin/checkpoint-200",
"epoch": 4.0,
"eval_steps": 200,
"global_step": 428,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.9345794392523364,
"grad_norm": 64164.80078125,
"learning_rate": 2.6031746031746032e-05,
"loss": 0.6523,
"step": 100
},
{
"epoch": 1.8691588785046729,
"grad_norm": 476353.5,
"learning_rate": 1.8095238095238094e-05,
"loss": 0.5671,
"step": 200
},
{
"epoch": 1.8691588785046729,
"eval_accuracy": 0.6589442815249267,
"eval_f1": 0.6579155363637312,
"eval_loss": 0.6257979869842529,
"eval_matthews_correlation": 0.3364903589596133,
"eval_precision": 0.6705934612651031,
"eval_recall": 0.6659292226576259,
"eval_runtime": 2.6504,
"eval_samples_per_second": 1286.606,
"eval_steps_per_second": 5.282,
"step": 200
},
{
"epoch": 2.803738317757009,
"grad_norm": 219894.8125,
"learning_rate": 1.0158730158730158e-05,
"loss": 0.3668,
"step": 300
},
{
"epoch": 3.7383177570093458,
"grad_norm": 366126.6875,
"learning_rate": 2.222222222222222e-06,
"loss": 0.1433,
"step": 400
},
{
"epoch": 3.7383177570093458,
"eval_accuracy": 0.6480938416422287,
"eval_f1": 0.6460041026152096,
"eval_loss": 1.1342097520828247,
"eval_matthews_correlation": 0.29208396951706195,
"eval_precision": 0.6462142217913729,
"eval_recall": 0.6458699506170023,
"eval_runtime": 2.8648,
"eval_samples_per_second": 1190.321,
"eval_steps_per_second": 4.887,
"step": 400
},
{
"epoch": 4.0,
"step": 428,
"total_flos": 2.69889186396186e+16,
"train_loss": 0.4096111087041481,
"train_runtime": 263.1821,
"train_samples_per_second": 414.542,
"train_steps_per_second": 1.626
}
],
"logging_steps": 100,
"max_steps": 428,
"num_input_tokens_seen": 0,
"num_train_epochs": 4,
"save_steps": 200,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 2.69889186396186e+16,
"train_batch_size": 64,
"trial_name": null,
"trial_params": null
}