Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    management terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     are
    0.57
     be
    0.54
    的
    0.50
    ளின்
    0.48
    king
    0.44
     ذلك
    0.44
    gebnisse
    0.44
    1
    0.44
     maliciously
    0.43
    는
    0.43
    POSITIVE LOGITS
    લ
    0.59
    ર
    0.57
    ल
    0.55
    ল
    0.55
    ל
    0.53
    ट
    0.52
    FBSDKLogin
    0.50
     expectations
    0.50
     entitles
    0.48
    ణ్
    0.48
    Act Density 0.020%

    No Known Activations