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

    Neuronpedia

    INDEX
    Explanations

    writing refusals

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Okay
    0.43
    Ok
    0.38
    rivation
    0.38
     المُ
    0.38
     चुनौतीपूर्ण
    0.36
    வு
    0.36
    跟大家
    0.36
    зах
    0.35
    ностями
    0.35
     fase
    0.35
    POSITIVE LOGITS
     rewrite
    0.45
     writing
    0.44
     wrote
    0.44
     schrij
    0.43
     write
    0.43
     escrib
    0.41
     пишет
    0.41
     লিখতে
    0.41
     писал
    0.41
     yazı
    0.41
    Act Density 0.000%

    No Known Activations