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
     Mate
    -0.08
    (['/
    -0.07
    rtc
    -0.07
     trên
    -0.07
    汇
    -0.07
    <|eom_id|>
    -0.07
    Greg
    -0.07
    ('_
    -0.07
     "*"
    -0.07
    .exceptions
    -0.06
    POSITIVE LOGITS
     उठ
    0.07
    му
    0.06
     진짜
    0.06
    ород
    0.06
    bourne
    0.06
    ológ
    0.06
    аниц
    0.06
     захисту
    0.06
    】【
    0.06
    -start
    0.05
    Act Density 0.003%

    No Known Activations