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

    Neuronpedia

    INDEX
    Explanations

    understanding or lack thereof

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    (sockfd
    -0.06
    lando
    -0.06
     mish
    -0.06
     Repos
    -0.06
     Combined
    -0.06
     administr
    -0.06
    .tom
    -0.06
     mostly
    -0.06
     /\
    -0.06
     doctor
    -0.06
    POSITIVE LOGITS
    ิ์
    0.07
    치를
    0.07
    ``,
    0.07
    agree
    0.06
     앞
    0.06
     Self
    0.06
     homosexuals
    0.06
    _Trans
    0.06
    trfs
    0.06
    Self
    0.06
    Act Density 0.030%

    No Known Activations