backdoored_bert-finetuned-sst2

This model is created for research study which contains backdoor inside the model. Please use it for academic research, don't use it for business scenarios.

This model is a fine-tuned version of Lujia/backdoored_bert on the glue dataset. It achieves the following results on the evaluation set:

• Loss: 0.4265
• Accuracy: 0.9255

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.1821	1.0	4210	0.2600	0.9151
0.1245	2.0	8420	0.3431	0.9197
0.0934	3.0	12630	0.3466	0.9186
0.0546	4.0	16840	0.3703	0.9232
0.0329	5.0	21050	0.4265	0.9255

Framework versions

• Transformers 4.27.4
• Pytorch 1.13.1+cu117
• Datasets 2.11.0
• Tokenizers 0.13.2