Add pipeline tag, library name and link to Github repository
Browse filesThis PR improves the model card by adding the pipeline_tag and library_name.
It ensures the model can be found at https://huggingface.co/models?pipeline_tag=text-generation and the how to use button is displayed.
README.md
CHANGED
|
@@ -1,13 +1,16 @@
|
|
| 1 |
---
|
| 2 |
-
license: mit
|
| 3 |
language:
|
| 4 |
- en
|
|
|
|
|
|
|
|
|
|
| 5 |
tags:
|
| 6 |
- llm
|
| 7 |
- safety
|
| 8 |
- jailbreak
|
| 9 |
- knowledge
|
| 10 |
---
|
|
|
|
| 11 |
# Introduction
|
| 12 |
|
| 13 |
This is a model for generating a jailbreak prompt based on knowledge point texts. The model is trained on the Llama-2-7b dataset and fine-tuned on the Knowledge-to-Jailbreak dataset. The model is intended to bridge the gap between theoretical vulnerabilities and real-world application scenarios, simulating sophisticated adversarial attacks that incorporate specialized knowledge.
|
|
@@ -228,7 +231,11 @@ max_tokens = 64
|
|
| 228 |
|
| 229 |
knowledge_points = ["Kettling Kettling (also known as containment or corralling) is a police tactic for controlling large crowds during demonstrations or protests. It involves the formation of large cordons of police officers who then move to contain a crowd within a limited area. Protesters are left only one choice of exit controlled by the police – or are completely prevented from leaving, with the effect of denying the protesters access to food, water and toilet facilities for a time period determined by the police forces. The tactic has proved controversial, in part because it has resulted in the detention of ordinary bystanders."]
|
| 230 |
|
| 231 |
-
batch_texts = [f'### Input:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 232 |
|
| 233 |
inputs = tokenizer(batch_texts, return_tensors='pt', padding=True, truncation=True, max_length=max_length - max_tokens).to(model.device)
|
| 234 |
|
|
@@ -246,6 +253,8 @@ print(generated_texts)
|
|
| 246 |
|
| 247 |
```
|
| 248 |
|
|
|
|
|
|
|
| 249 |
# Citation
|
| 250 |
|
| 251 |
If you find this model useful, please cite the following paper:
|
|
|
|
| 1 |
---
|
|
|
|
| 2 |
language:
|
| 3 |
- en
|
| 4 |
+
license: mit
|
| 5 |
+
pipeline_tag: text-generation
|
| 6 |
+
library_name: transformers
|
| 7 |
tags:
|
| 8 |
- llm
|
| 9 |
- safety
|
| 10 |
- jailbreak
|
| 11 |
- knowledge
|
| 12 |
---
|
| 13 |
+
|
| 14 |
# Introduction
|
| 15 |
|
| 16 |
This is a model for generating a jailbreak prompt based on knowledge point texts. The model is trained on the Llama-2-7b dataset and fine-tuned on the Knowledge-to-Jailbreak dataset. The model is intended to bridge the gap between theoretical vulnerabilities and real-world application scenarios, simulating sophisticated adversarial attacks that incorporate specialized knowledge.
|
|
|
|
| 231 |
|
| 232 |
knowledge_points = ["Kettling Kettling (also known as containment or corralling) is a police tactic for controlling large crowds during demonstrations or protests. It involves the formation of large cordons of police officers who then move to contain a crowd within a limited area. Protesters are left only one choice of exit controlled by the police – or are completely prevented from leaving, with the effect of denying the protesters access to food, water and toilet facilities for a time period determined by the police forces. The tactic has proved controversial, in part because it has resulted in the detention of ordinary bystanders."]
|
| 233 |
|
| 234 |
+
batch_texts = [f'### Input:
|
| 235 |
+
{input_}
|
| 236 |
+
|
| 237 |
+
### Response:
|
| 238 |
+
' for input_ in knowledge_points]
|
| 239 |
|
| 240 |
inputs = tokenizer(batch_texts, return_tensors='pt', padding=True, truncation=True, max_length=max_length - max_tokens).to(model.device)
|
| 241 |
|
|
|
|
| 253 |
|
| 254 |
```
|
| 255 |
|
| 256 |
+
Code for this and the datasets is available at https://github.com/THU-KEG/Knowledge-to-JailBreak.
|
| 257 |
+
|
| 258 |
# Citation
|
| 259 |
|
| 260 |
If you find this model useful, please cite the following paper:
|