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

    Neuronpedia

    INDEX
    Explanations

    Math typesetting

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     heartfelt
    -0.08
     translate
    -0.07
     reign
    -0.07
    .Translate
    -0.07
     Attached
    -0.07
    _ATTACH
    -0.07
     translating
    -0.07
     everyone's
    -0.07
     Insets
    -0.07
     clinically
    -0.07
    POSITIVE LOGITS
     unconventional
    0.09
     pédagog
    0.08
     invál
    0.08
    ębior
    0.08
     बनने
    0.08
    不是
    0.08
     बनाने
    0.08
    uč
    0.08
    äufig
    0.08
     हु�
    0.08
    Act Density 0.021%

    No Known Activations