© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 2027
    Prev
    Next
    INDEX
    Explanations

    explicit content requests

    np_acts-logits-general · gemini-2.5-flash-lite

    requests containing explicit sexual content or exploitative themes and hateful slurs—i.e., unsafe, policy-violating prompts.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     বিবেচনায়
    0.48
     reporting
    0.40
     forecast
    0.38
    ↵
    0.38
    对
    0.38
     Assessing
    0.37
     assessment
    0.37
     Reporting
    0.37
     typically
    0.37
    From
    0.37
    POSITIVE LOGITS
     ainult
    0.48
     HOWEVER
    0.46
     Sachen
    0.45
     terkenal
    0.44
    됬
    0.44
     blatantly
    0.43
    жалуйста
    0.43
     horrible
    0.43
     recomiendo
    0.43
     Pikachu
    0.43
    Activations Density 0.221%

    No Known Activations