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

    Neuronpedia

    INDEX
    Explanations

    ROTC

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    好吃
    -0.08
    禁忌
    -0.07
     robbery
    -0.07
     מעולה
    -0.07
    dar
    -0.07
    奇怪
    -0.07
    叹了口气
    -0.07
    さまざ
    -0.07
    مطل
    -0.07
    半月
    -0.07
    POSITIVE LOGITS
     some
    0.07
    Parsing
    0.07
     FY
    0.07
    ories
    0.07
    irement
    0.07
    -era
    0.06
    线下
    0.06
     الى
    0.06
    (j
    0.06
     enums
    0.06
    Act Density 0.005%

    No Known Activations