# Stealthiness Evaluation on LLaMA Guard 7B Source: [Stealthiness Evaluation on LLaMA Guard 7B]() Collection index: [llms.txt]() Data updated: 2026-05-18T09:28:34.687Z Evaluation of model stealthiness performance on the LLaMA Guard 7B dataset, measured using Perplexity (PPL) 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 | Stealthiness Evaluation | | Dataset (including any stated split) | LLaMA Guard 7B | | Metric | Mean Perplexity (PPL) | | Unit | SCALAR | | Direction | Lower is better | | Results loaded | 10 of 10 | | 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 - [Mean Perplexity (PPL)](): Lower is better; unit: SCALAR. - [Std Dev Perplexity (PPL)](): Lower is better; unit: SCALAR. ## Recorded results | Method | Mean Perplexity (PPL) | Result date | Report paper | | --- | --- | --- | --- | | Ascii | 3.05 | Not recorded | 68fa751a1e209131cea65ca8 | | ArtPrompt | 3.36 | Not recorded | 68fa751a1e209131cea65ca8 | | Base64 | 10.14 | Not recorded | 68fa751a1e209131cea65ca8 | | Morse Cipher | 11.18 | Not recorded | 68fa751a1e209131cea65ca8 | | ReNeLLM | 13.33 | Not recorded | 68fa751a1e209131cea65ca8 | | Unicode | 29.37 | Not recorded | 68fa751a1e209131cea65ca8 | | UTF-8 | 29.37 | Not recorded | 68fa751a1e209131cea65ca8 | | Origin | 33.44 | Not recorded | 68fa751a1e209131cea65ca8 | | Caesar Cipher | 202.08 | Not recorded | 68fa751a1e209131cea65ca8 | | FlipAttack | 563.32 | Not recorded | 68fa751a1e209131cea65ca8 | ### Ascii Mean Perplexity (PPL): 3.05. Result ID: 6a0adb6a47f34ce924f14ee1. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ArtPrompt Mean Perplexity (PPL): 3.36. Result ID: 6a0adb6a47f34ce924f14ee3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Base64 Mean Perplexity (PPL): 10.14. Result ID: 6a0adb6a47f34ce924f14ee2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Morse Cipher Mean Perplexity (PPL): 11.18. Result ID: 6a0adb6a47f34ce924f14edf. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ReNeLLM Mean Perplexity (PPL): 13.33. Result ID: 6a0adb6a47f34ce924f14ee4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Unicode Mean Perplexity (PPL): 29.37. Result ID: 6a0adb6a47f34ce924f14ede. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### UTF-8 Mean Perplexity (PPL): 29.37. Result ID: 6a0adb6a47f34ce924f14ee0. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Origin Mean Perplexity (PPL): 33.44. Result ID: 6a0adb6a47f34ce924f14edc. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Caesar Cipher Mean Perplexity (PPL): 202.08. Result ID: 6a0adb6a47f34ce924f14edd. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### FlipAttack Mean Perplexity (PPL): 563.32. Result ID: 6a0adb6a47f34ce924f14ee5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.