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

    Neuronpedia

    INDEX
    Explanations

    routing and modules

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    aniya
    0.54
    stackoverflow
    0.54
    ooky
    0.53
    azaki
    0.52
     cahaya
    0.52
     కంప
    0.52
    unggu
    0.51
    所有人
    0.51
    all
    0.51
     টেব
    0.51
    POSITIVE LOGITS
    ン
    0.68
    的
    0.59
     renforc
    0.54
    化
    0.53
     concerto
    0.52
    ﻫ
    0.52
     scor
    0.52
    プロジェクト
    0.52
    R
    0.52
    ベル
    0.51
    Act Density 0.006%

    No Known Activations