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

    Neuronpedia

    INDEX
    Explanations

    median and markdown format

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    स
    1.02
    д
    0.84
     as
    0.76
    ہ
    0.76
    ков
    0.75
    یت
    0.73
    kannya
    0.71
     algebras
    0.70
    𝓭
    0.68
    सं
    0.68
    POSITIVE LOGITS
    et
    1.19
    a
    1.17
    on
    1.06
    il
    0.98
    in
    0.97
    া
    0.96
    ר
    0.92
    ี
    0.88
    ر
    0.86
    r
    0.85
    Act Density 0.026%

    No Known Activations