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

    Neuronpedia

    INDEX
    Explanations

    a model specific

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ご飯
    1.66
    rary
    1.46
    に向け
    1.40
    لوب
    1.36
    全体
    1.34
    しゃれ
    1.31
     both
    1.30
     other
    1.29
     either
    1.28
    狽
    1.28
    POSITIVE LOGITS
    HC
    2.31
    WS
    2.28
    BC
    2.27
    CP
    2.25
    ND
    2.23
    JP
    2.21
    bc
    2.20
    MM
    2.19
    DK
    2.19
    HH
    2.19
    Act Density 0.647%

    No Known Activations