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

    Neuronpedia

    INDEX
    Explanations

    type

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    .clock
    -0.07
     breeding
    -0.07
    recv
    -0.07
    plt
    -0.07
    .cor
    -0.07
     Polo
    -0.07
    (socket
    -0.06
     Jeg
    -0.06
     Slovakia
    -0.06
     Shakespeare
    -0.06
    POSITIVE LOGITS
    ARCH
    0.08
     uploader
    0.07
     generics
    0.07
     CRA
    0.07
     textures
    0.06
     SX
    0.06
     UD
    0.06
    שמח
    0.06
    行く
    0.06
     trope
    0.06
    Act Density 0.002%

    No Known Activations