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

    Neuronpedia

    INDEX
    Explanations

    Okay, let's analyze

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    il
    1.25
    ce
    1.25
    en
    1.20
    يت
    1.17
    ere
    1.15
    ets
    1.14
    c
    1.10
    es
    1.09
    có
    1.05
    o
    1.04
    POSITIVE LOGITS
     aquesta
    1.20
     Règles
    1.20
     allemande
    1.13
     Oké
    1.13
    𝐭
    1.12
     infert
    1.09
     renverse
    1.09
     januar
    1.08
     MACHINIST
    1.06
    션
    1.06
    Act Density 0.217%

    No Known Activations