© 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. 84421
    Prev
    Next
    INDEX
    Explanations

    escaping from danger or confinement

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

    escaping from situations

    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
    е
    0.73
    '
    0.71
    บริหาร
    0.58
    ત
    0.57
    ,”
    0.56
    iyê
    0.55
    ’
    0.55
    였습니다
    0.53
    ”,
    0.52
     맞춰
    0.52
    POSITIVE LOGITS
     escapes
    1.34
     escape
    1.31
     escaping
    1.26
     escaped
    1.23
    escape
    1.22
     Escape
    1.17
    Escape
    1.13
    逃
    1.11
    escaping
    1.09
     escapar
    1.01
    Activations Density 0.047%

    No Known Activations