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

    Neuronpedia

    INDEX
    Explanations

    non-English or unusual punctuation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    elope
    -0.07
    ゃ
    -0.06
    ProgressHUD
    -0.06
     металли
    -0.06
     кус
    -0.06
     taraf
    -0.06
    alez
    -0.06
    рус
    -0.06
    icer
    -0.06
    Nej
    -0.06
    POSITIVE LOGITS
    ---@
    0.07
    Emer
    0.07
     während
    0.07
     estruct
    0.07
    (enum
    0.06
    いつ
    0.06
    COMM
    0.06
     plug
    0.06
    .reason
    0.06
     extremism
    0.06
    Act Density 0.015%

    No Known Activations