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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     behavioral
    -0.08
    .dest
    -0.07
    ്
    -0.07
    FONT
    -0.06
    EXPECTED
    -0.06
    ition
    -0.06
    <App
    -0.06
     kriz
    -0.06
     Hutchinson
    -0.06
     deployment
    -0.06
    POSITIVE LOGITS
    etype
    0.06
    _original
    0.06
    ľ
    0.06
     dood
    0.06
     Fucked
    0.06
    麼
    0.06
    吸
    0.06
     ade
    0.06
    (separator
    0.05
    áte
    0.05
    Act Density 0.040%

    No Known Activations