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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 1844
    Prev
    Next
    INDEX
    Explanations

    avoid, reject, deny, evade

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

    avoiding complex or inappropriate content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     этом
    0.27
     prieš
    0.26
     cubre
    0.26
     þei
    0.25
     ihnen
    0.25
     어려
    0.25
    猕
    0.25
    ),\\
    0.25
     njima
    0.24
     chậm
    0.24
    POSITIVE LOGITS
    the
    0.31
    ell
    0.30
    s
    0.30
    ,
    0.27
    OL
    0.27
     hometown
    0.27
    .
    0.27
    il
    0.26
    :
    0.26
    ul
    0.26
    Activations Density 0.262%

    No Known Activations