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

    Neuronpedia

    INDEX
    Explanations

    Okay, let's break down

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     नहीं
    0.33
    㫧
    0.32
    இதனால்
    0.31
     সতর্কতা
    0.31
    秞
    0.31
    ждую
    0.30
     まま
    0.30
     ഒഴിവാ
    0.29
    ޏ
    0.29
    😐
    0.29
    POSITIVE LOGITS
     extensively
    0.33
     The
    0.32
    highly
    0.31
    The
    0.30
    izu
    0.30
     intricate
    0.29
    performing
    0.29
    we
    0.28
    extensive
    0.28
     This
    0.28
    Act Density 0.187%

    No Known Activations