# Jailbreak Attack on In-the-wild Jailbreak Source: [Jailbreak Attack on In-the-wild Jailbreak]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:24:03.903Z Evaluation of model performance against adversarial jailbreak prompts using the In-the-wild Jailbreak dataset, measured by various Attack Success Rate (ASR) metrics. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Jailbreak Attack | | Dataset (including any stated split) | In-the-wild Jailbreak | | Metric | ASR (m) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 15 of 15 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [ASR (m)](): Higher is better; unit: PERCENT. - [ASR](): Higher is better; unit: PERCENT. ## Recorded results | Method | ASR (m) | Result date | Report paper | | --- | --- | --- | --- | | S-PIEL | 60.3 | Not recorded | 68fa7c001e209131cea8bd0d | | S-PIEL | 58.7 | Not recorded | 68fa7c001e209131cea8bd0d | | S-PIEL | 54.3 | Not recorded | 68fa7c001e209131cea8bd0d | | S-PIEL | 38.9 | Not recorded | 68fa7c001e209131cea8bd0d | | S-PIEL | 36.9 | Not recorded | 68fa7c001e209131cea8bd0d | | GCG | 19.7 | Not recorded | 68fa7c001e209131cea8bd0d | | GCG | 18.9 | Not recorded | 68fa7c001e209131cea8bd0d | | GCG | 18.8 | Not recorded | 68fa7c001e209131cea8bd0d | | Vanilla Prompt | 18.7 | Not recorded | 68fa7c001e209131cea8bd0d | | Vanilla Prompt | 12.3 | Not recorded | 68fa7c001e209131cea8bd0d | | GCG | 12.2 | Not recorded | 68fa7c001e209131cea8bd0d | | Vanilla Prompt | 12.1 | Not recorded | 68fa7c001e209131cea8bd0d | | GCG | 8.9 | Not recorded | 68fa7c001e209131cea8bd0d | | Vanilla Prompt | 7.7 | Not recorded | 68fa7c001e209131cea8bd0d | | Vanilla Prompt | 6.5 | Not recorded | 68fa7c001e209131cea8bd0d | ### S-PIEL ASR (m): 60.3. Result ID: 698b8dee7a967eddb76f89ee. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model Type":"Mistral 7B"}. This record does not assert independent reproduction. ### S-PIEL ASR (m): 58.7. Result ID: 698b8dee7a967eddb76f89f1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model Type":"Gemma-2-9B"}. This record does not assert independent reproduction. ### S-PIEL ASR (m): 54.3. Result ID: 698b8dee7a967eddb76f89e8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model Type":"Llama-2-7B"}. This record does not assert independent reproduction. ### S-PIEL ASR (m): 38.9. Result ID: 698b8dee7a967eddb76f89eb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model Type":"Llama-3.1-8B"}. This record does not assert independent reproduction. ### S-PIEL ASR (m): 36.9. Result ID: 698b8dee7a967eddb76f89f4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model Type":"Llama-3.1-8B (DeepSeek-R1-Distilled)"}. This record does not assert independent reproduction. ### GCG ASR (m): 19.7. Result ID: 698b8dee7a967eddb76f89ed. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model Type":"Mistral 7B"}. This record does not assert independent reproduction. ### GCG ASR (m): 18.9. Result ID: 698b8dee7a967eddb76f89e7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model Type":"Llama-2-7B"}. This record does not assert independent reproduction. ### GCG ASR (m): 18.8. Result ID: 698b8dee7a967eddb76f89f0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model Type":"Gemma-2-9B"}. This record does not assert independent reproduction. ### Vanilla Prompt ASR (m): 18.7. Result ID: 698b8dee7a967eddb76f89ec. [Reporting paper]() Recorded evaluation configuration: {"Target Model Type":"Mistral 7B"}. This record does not assert independent reproduction. ### Vanilla Prompt ASR (m): 12.3. Result ID: 698b8dee7a967eddb76f89ef. [Reporting paper]() Recorded evaluation configuration: {"Target Model Type":"Gemma-2-9B"}. This record does not assert independent reproduction. ### GCG ASR (m): 12.2. Result ID: 698b8dee7a967eddb76f89ea. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model Type":"Llama-3.1-8B"}. This record does not assert independent reproduction. ### Vanilla Prompt ASR (m): 12.1. Result ID: 698b8dee7a967eddb76f89e6. [Reporting paper]() Recorded evaluation configuration: {"Target Model Type":"Llama-2-7B"}. This record does not assert independent reproduction. ### GCG ASR (m): 8.9. Result ID: 698b8dee7a967eddb76f89f3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model Type":"Llama-3.1-8B (DeepSeek-R1-Distilled)"}. This record does not assert independent reproduction. ### Vanilla Prompt ASR (m): 7.7. Result ID: 698b8dee7a967eddb76f89e9. [Reporting paper]() Recorded evaluation configuration: {"Target Model Type":"Llama-3.1-8B"}. This record does not assert independent reproduction. ### Vanilla Prompt ASR (m): 6.5. Result ID: 698b8dee7a967eddb76f89f2. [Reporting paper]() Recorded evaluation configuration: {"Target Model Type":"Llama-3.1-8B (DeepSeek-R1-Distilled)"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.