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

    Neuronpedia

    INDEX
    Explanations

    word count instructions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     giorni
    0.42
    plicht
    0.39
     atribut
    0.37
     exemplaire
    0.36
    ченко
    0.36
    Worked
    0.36
    дьми
    0.35
    вні
    0.35
    ើរ
    0.35
    Encript
    0.35
    POSITIVE LOGITS
    +
    0.38
     `
    0.36
    aces
    0.34
    cing
    0.32
    .
    0.32
    👋
    0.32
     French
    0.31
     https
    0.31
    中使用
    0.30
     Setter
    0.29
    Act Density 0.028%

    No Known Activations