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

    Neuronpedia

    INDEX
    Explanations

    directions east and northeast

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    m
    1.55
    м
    1.22
    م
    1.22
    _
    1.17
    N
    1.13
    L
    1.09
    am
    1.06
    ur
    1.04
    '
    1.01
    로
    1.01
    POSITIVE LOGITS
    та
    1.06
     to
    1.03
    的同时
    1.01
     defens
    0.95
    ли
    0.93
    σ
    0.91
    *,
    0.89
    ,
    0.88
     oxide
    0.88
    的产品
    0.83
    Act Density 0.001%

    No Known Activations