learning_rate: 2e-05
train_batch_size: 16
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 300
num_epochs: 6
label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.4426	0.5556	500	0.4444	0.9233	0.9236	0.9260	0.9233
0.3827	1.1111	1000	0.3956	0.9440	0.9439	0.9439	0.9440
0.3837	1.6667	1500	0.4054	0.9451	0.9454	0.9465	0.9451
0.3314	2.2222	2000	0.3659	0.9626	0.9627	0.9629	0.9626
0.3354	2.7778	2500	0.3798	0.9625	0.9627	0.9635	0.9625
0.3201	3.3333	3000	0.3804	0.9632	0.9632	0.9637	0.9632
0.3122	3.8889	3500	0.3716	0.9682	0.9682	0.9682	0.9682
0.3005	4.4444	4000	0.3713	0.9692	0.9692	0.9694	0.9692
0.3022	5.0	4500	0.3677	0.9708	0.9708	0.9709	0.9708
0.3022	5.5556	5000	0.3673	0.9704	0.9705	0.9705	0.9704

Framework versions

Transformers 4.51.3
Pytorch 2.1.2
Datasets 3.5.1
Tokenizers 0.21.1

Training info:

TrainOutput(global_step=5400, training_loss=0.3579492653740777, metrics={'train_runtime': 7336.8949, 'train_samples_per_second': 23.552, 'train_steps_per_second': 0.736, 'total_flos': 4.54659985833984e+16, 'train_loss': 0.3579492653740777, 'epoch': 6.0})

Ghazi-nak/8_BiasDetctor_HTW_ANG.KI

8_BiasDetctor_HTW_ANG.KI

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Training info:

Ghazi-nak/
8_BiasDetctor_HTW_ANG.KI