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
     errone
    -0.07
    ニー
    -0.07
    .o
    -0.07
     (:
    -0.06
     ولی
    -0.06
     chickens
    -0.06
    tracked
    -0.06
    _normalize
    -0.06
     Роз
    -0.06
    }:
    -0.06
    POSITIVE LOGITS
     nokt
    0.06
     citt
    0.06
    ẩm
    0.06
    亞
    0.06
    removeClass
    0.06
    odal
    0.05
    лиж
    0.05
    .integer
    0.05
     Sierra
    0.05
    亚
    0.05
    Act Density 0.002%

    No Known Activations