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

    Neuronpedia

    INDEX
    Explanations

    FC/AFC

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Couldn
    -0.07
    ]byte
    -0.06
    \u
    -0.06
    read
    -0.06
     براي
    -0.06
     pred
    -0.06
    iter
    -0.06
     cosine
    -0.06
    _curr
    -0.06
    ць
    -0.06
    POSITIVE LOGITS
     anarchists
    0.07
    FTA
    0.07
    聊
    0.07
    fang
    0.06
     fencing
    0.06
     NFC
    0.06
    _finder
    0.06
    _slot
    0.06
    �인
    0.06
     showModal
    0.06
    Act Density 0.013%

    No Known Activations