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

    Neuronpedia

    INDEX
    Explanations

    Casual, Collabor, Standard, Covid, OH, Customer

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     at
    1.60
     of
    1.53
     was
    1.53
     to
    1.40
    y
    1.27
     with
    1.22
    h
    1.20
    m
    1.20
     by
    1.16
     been
    1.13
    POSITIVE LOGITS
    ও
    1.63
    在
    1.46
    .
    1.39
    도
    1.29
    ка
    1.24
    も
    1.20
    出
    1.15
    ↵
    1.13
     في
    1.11
    高
    1.09
    Act Density 0.059%

    No Known Activations