RoBERTa-10 / checkpoint-750 /trainer_state.json
dzd828's picture
Upload folder using huggingface_hub
4a3ca35 verified
{
"best_metric": 2.0163416862487793,
"best_model_checkpoint": "910-3-roberta-10/checkpoint-750",
"epoch": 3.0,
"eval_steps": 500,
"global_step": 750,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.1,
"grad_norm": 23.742645263671875,
"learning_rate": 1.6666666666666667e-06,
"loss": 2.3895,
"step": 25
},
{
"epoch": 0.2,
"grad_norm": 11.274296760559082,
"learning_rate": 3.3333333333333333e-06,
"loss": 2.3348,
"step": 50
},
{
"epoch": 0.3,
"grad_norm": 7.359958171844482,
"learning_rate": 5e-06,
"loss": 2.3217,
"step": 75
},
{
"epoch": 0.4,
"grad_norm": 8.179986000061035,
"learning_rate": 6.666666666666667e-06,
"loss": 2.3207,
"step": 100
},
{
"epoch": 0.5,
"grad_norm": 18.304018020629883,
"learning_rate": 8.333333333333334e-06,
"loss": 2.2965,
"step": 125
},
{
"epoch": 0.6,
"grad_norm": 12.443976402282715,
"learning_rate": 1e-05,
"loss": 2.3036,
"step": 150
},
{
"epoch": 0.7,
"grad_norm": 14.003621101379395,
"learning_rate": 1.1666666666666668e-05,
"loss": 2.2321,
"step": 175
},
{
"epoch": 0.8,
"grad_norm": 12.123735427856445,
"learning_rate": 1.3333333333333333e-05,
"loss": 2.218,
"step": 200
},
{
"epoch": 0.9,
"grad_norm": 26.824359893798828,
"learning_rate": 1.5e-05,
"loss": 2.1573,
"step": 225
},
{
"epoch": 1.0,
"grad_norm": 12.286924362182617,
"learning_rate": 1.6666666666666667e-05,
"loss": 2.146,
"step": 250
},
{
"epoch": 1.0,
"eval_accuracy": 0.1845,
"eval_f1_macro": 0.0987979652095434,
"eval_f1_micro": 0.1845,
"eval_f1_weighted": 0.09879796520954343,
"eval_loss": 2.1069414615631104,
"eval_precision_macro": 0.10847279767673075,
"eval_precision_micro": 0.1845,
"eval_precision_weighted": 0.10847279767673075,
"eval_recall_macro": 0.1845,
"eval_recall_micro": 0.1845,
"eval_recall_weighted": 0.1845,
"eval_runtime": 5.8665,
"eval_samples_per_second": 340.92,
"eval_steps_per_second": 5.455,
"step": 250
},
{
"epoch": 1.1,
"grad_norm": 13.818865776062012,
"learning_rate": 1.8333333333333333e-05,
"loss": 2.1453,
"step": 275
},
{
"epoch": 1.2,
"grad_norm": 16.149505615234375,
"learning_rate": 2e-05,
"loss": 2.1011,
"step": 300
},
{
"epoch": 1.3,
"grad_norm": 24.287403106689453,
"learning_rate": 2.1666666666666667e-05,
"loss": 2.087,
"step": 325
},
{
"epoch": 1.4,
"grad_norm": 8.6444673538208,
"learning_rate": 2.3333333333333336e-05,
"loss": 2.11,
"step": 350
},
{
"epoch": 1.5,
"grad_norm": 18.66366195678711,
"learning_rate": 2.5e-05,
"loss": 2.0991,
"step": 375
},
{
"epoch": 1.6,
"grad_norm": 8.785151481628418,
"learning_rate": 2.6666666666666667e-05,
"loss": 2.0281,
"step": 400
},
{
"epoch": 1.7,
"grad_norm": 9.933465957641602,
"learning_rate": 2.8333333333333335e-05,
"loss": 2.0634,
"step": 425
},
{
"epoch": 1.8,
"grad_norm": 13.846152305603027,
"learning_rate": 3e-05,
"loss": 2.0504,
"step": 450
},
{
"epoch": 1.9,
"grad_norm": 24.698633193969727,
"learning_rate": 3.1666666666666666e-05,
"loss": 2.0754,
"step": 475
},
{
"epoch": 2.0,
"grad_norm": 13.607393264770508,
"learning_rate": 3.3333333333333335e-05,
"loss": 2.0,
"step": 500
},
{
"epoch": 2.0,
"eval_accuracy": 0.224,
"eval_f1_macro": 0.15699864365105617,
"eval_f1_micro": 0.224,
"eval_f1_weighted": 0.15699864365105617,
"eval_loss": 2.0973894596099854,
"eval_precision_macro": 0.17293733079797044,
"eval_precision_micro": 0.224,
"eval_precision_weighted": 0.17293733079797044,
"eval_recall_macro": 0.22400000000000003,
"eval_recall_micro": 0.224,
"eval_recall_weighted": 0.224,
"eval_runtime": 5.9646,
"eval_samples_per_second": 335.311,
"eval_steps_per_second": 5.365,
"step": 500
},
{
"epoch": 2.1,
"grad_norm": 10.633151054382324,
"learning_rate": 3.5e-05,
"loss": 1.9934,
"step": 525
},
{
"epoch": 2.2,
"grad_norm": 11.844949722290039,
"learning_rate": 3.6666666666666666e-05,
"loss": 1.9635,
"step": 550
},
{
"epoch": 2.3,
"grad_norm": 11.696029663085938,
"learning_rate": 3.8333333333333334e-05,
"loss": 1.9717,
"step": 575
},
{
"epoch": 2.4,
"grad_norm": 10.775115966796875,
"learning_rate": 4e-05,
"loss": 1.99,
"step": 600
},
{
"epoch": 2.5,
"grad_norm": 12.655854225158691,
"learning_rate": 4.166666666666667e-05,
"loss": 1.9839,
"step": 625
},
{
"epoch": 2.6,
"grad_norm": 11.810218811035156,
"learning_rate": 4.3333333333333334e-05,
"loss": 1.9982,
"step": 650
},
{
"epoch": 2.7,
"grad_norm": 8.966118812561035,
"learning_rate": 4.5e-05,
"loss": 1.9561,
"step": 675
},
{
"epoch": 2.8,
"grad_norm": 14.116168975830078,
"learning_rate": 4.666666666666667e-05,
"loss": 1.9757,
"step": 700
},
{
"epoch": 2.9,
"grad_norm": 26.801786422729492,
"learning_rate": 4.8333333333333334e-05,
"loss": 2.0137,
"step": 725
},
{
"epoch": 3.0,
"grad_norm": 21.48545265197754,
"learning_rate": 5e-05,
"loss": 2.0136,
"step": 750
},
{
"epoch": 3.0,
"eval_accuracy": 0.234,
"eval_f1_macro": 0.17560544673032039,
"eval_f1_micro": 0.234,
"eval_f1_weighted": 0.17560544673032036,
"eval_loss": 2.0163416862487793,
"eval_precision_macro": 0.3230402136577736,
"eval_precision_micro": 0.234,
"eval_precision_weighted": 0.3230402136577736,
"eval_recall_macro": 0.23400000000000004,
"eval_recall_micro": 0.234,
"eval_recall_weighted": 0.234,
"eval_runtime": 5.9722,
"eval_samples_per_second": 334.883,
"eval_steps_per_second": 5.358,
"step": 750
}
],
"logging_steps": 25,
"max_steps": 7500,
"num_input_tokens_seen": 0,
"num_train_epochs": 30,
"save_steps": 500,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 5,
"early_stopping_threshold": 0.01
},
"attributes": {
"early_stopping_patience_counter": 0
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 5591739389706240.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}