Update README.md
Browse files
README.md
CHANGED
|
@@ -5,6 +5,18 @@ language:
|
|
| 5 |
base_model:
|
| 6 |
- meta-llama/Prompt-Guard-86M
|
| 7 |
pipeline_tag: text-classification
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
---
|
| 9 |
# katanemo/Arch-Guard
|
| 10 |
|
|
@@ -29,22 +41,7 @@ In summary, the Katanemo Arch-Function collection demonstrates:
|
|
| 29 |
The gpu model is quantized with EEtq, please follow the instruction at https://github.com/NetEase-FuXi/EETQ?tab=readme-ov-file#getting-started to install the package.
|
| 30 |
|
| 31 |
## Datasets
|
| 32 |
-
Evaluation dataset is from
|
| 33 |
-
[casual_conversation](https://huggingface.co/datasets/SohamGhadge/casual-conversation)
|
| 34 |
-
[commonqa](https://huggingface.co/datasets/tau/commonsense_qa)
|
| 35 |
-
[financeqa](https://huggingface.co/datasets/AIR-Bench/qa_finance_en)
|
| 36 |
-
[instruction](http://mbzuai/LaMini-instruction)
|
| 37 |
-
[jailbreak_behavior_benign](https://huggingface.co/datasets/JailbreakBench/JBB-Behaviors)
|
| 38 |
-
[jailbreak_behavior_harmful](https://huggingface.co/datasets/JailbreakBench/JBB-Behaviors)
|
| 39 |
-
[jailbreak_judge](https://huggingface.co/datasets/JailbreakBench/JBB-Behaviors)
|
| 40 |
-
[jailbreak_prompts](https://huggingface.co/datasets/rubend18/ChatGPT-Jailbreak-Prompts)
|
| 41 |
-
[jailbreak_tweet](https://huggingface.co/datasets/cstnz/Disaster-tweet-jailbreaking)
|
| 42 |
-
[jailbreak_v](https://huggingface.co/datasets/JailbreakV-28K/JailBreakV-28k)
|
| 43 |
-
[jailbreak_vigil](https://huggingface.co/datasets/deadbits/vigil-jailbreak-all-MiniLM-L6-v2)
|
| 44 |
-
[mental_health](https://huggingface.co/datasets/Amod/mental_health_counseling_conversations)
|
| 45 |
-
[telecom](https://huggingface.co/datasets/talkmap/telecom-conversation-corpus)
|
| 46 |
-
[truthqa](https://huggingface.co/datasets/truthfulqa/truthful_qa)
|
| 47 |
-
[weather](https://huggingface.co/datasets/GEM/conversational_weather)
|
| 48 |
|
| 49 |
## How to use
|
| 50 |
|
|
|
|
| 5 |
base_model:
|
| 6 |
- meta-llama/Prompt-Guard-86M
|
| 7 |
pipeline_tag: text-classification
|
| 8 |
+
datasets:
|
| 9 |
+
- SohamGhadge/casual-conversation
|
| 10 |
+
- tau/commonsense_qa
|
| 11 |
+
- AIR-Bench/qa_finance_en
|
| 12 |
+
- JailbreakBench/JBB-Behaviors
|
| 13 |
+
- rubend18/ChatGPT-Jailbreak-Prompts
|
| 14 |
+
- cstnz/Disaster-tweet-jailbreaking
|
| 15 |
+
- JailbreakV-28K/JailBreakV-28k
|
| 16 |
+
- Amod/mental_health_counseling_conversations
|
| 17 |
+
- talkmap/telecom-conversation-corpus
|
| 18 |
+
- truthfulqa/truthful_qa
|
| 19 |
+
- GEM/conversational_weather
|
| 20 |
---
|
| 21 |
# katanemo/Arch-Guard
|
| 22 |
|
|
|
|
| 41 |
The gpu model is quantized with EEtq, please follow the instruction at https://github.com/NetEase-FuXi/EETQ?tab=readme-ov-file#getting-started to install the package.
|
| 42 |
|
| 43 |
## Datasets
|
| 44 |
+
Evaluation dataset is sourced from a combination of open source datasets.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
|
| 46 |
## How to use
|
| 47 |
|