tom-gibbs/multi-turn_jailbreak_attack_datasets
Multi-Turn Jailbreak Attack Datasets Description This dataset was created to compare single-turn and multi-turn jailbreak attacks on large language models (LLMs). The primary goal is to take a single harmful prompt and distribute the harm over multiple turns, making each prompt appear harmless in isolation. This approach is compared against traditional single-turn attacks with the complete prompt to understand their relative impacts and failure modes. The key… See the full description on the dataset page: https://huggingface.co/datasets/tom-gibbs/multi-turn_jailbreak_attack_datasets.
| Author / organization | tom-gibbs |
|---|---|
| License (as declared) | mit |
| Downloads | 1.1K (as reported by Hugging Face) |
| Likes | 20 |
| Created | 18 August 2024 |
| Last updated at source | 7 September 2024 |
Popularity, last 3 days
1,093 now (+4 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 17 Oct 2026 Description: Multi-Turn Jailbreak Attack Datasets Description This dataset was created to compare single-turn and multi-turn jailbreak attacks on large language models (LLMs). The primary goal is to take a single harmful prompt and distribute the harm o → Multi-Turn Jailbreak Attack Datasets Description This dataset was created to compare single-turn and multi-turn jailbreak attacks on large language models (LLMs). The primary goal is to take a single harmful prompt and distribute the harm o
Changes detected by Silytics between refreshes of the source.
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.