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

    Neuronpedia

    INDEX
    Explanations

    application enters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     заме
    1.08
    ன்
    1.06
     रव
    1.02
    sop
    1.01
    न
    1.01
    sword
    0.96
    以
    0.93
    Ѕ
    0.92
    t
    0.89
    ス
    0.89
    POSITIVE LOGITS
    ,\
    1.31
    ﻯ
    1.21
    impanan
    1.19
    是要
    1.13
    upaten
    1.09
    てください
    1.08
     речь
    1.07
    1.06
    thedocs
    1.05
     effectuer
    1.04
    Act Density 0.002%

    No Known Activations