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

    Neuronpedia

    INDEX
    Explanations

    breakdown of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ba
    1.12
    𝓁
    1.09
    𝒽
    1.09
    ।''
    1.08
    ﻭ
    1.08
    luğu
    1.07
    begin
    1.05
    𝓈
    1.03
    sc
    1.02
    𝒹
    1.02
    POSITIVE LOGITS
    ين
    1.52
    其他
    1.29
    ز
    1.13
    ють
    1.09
    ال
    1.07
    да
    1.06
    д
    1.03
     elettron
    1.02
     rosette
    0.97
    อง
    0.96
    Act Density 0.066%

    No Known Activations