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

    Neuronpedia

    INDEX
    Explanations

    say, incorrect responses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    tion
    1.21
    ्
    1.17
    y
    1.07
    tions
    1.04
    eek
    1.02
    entuk
    1.01
    t
    1.00
    اگ
    0.96
     locust
    0.95
     cavitation
    0.95
    POSITIVE LOGITS
    스의
    1.25
    を満
    1.18
    스를
    1.15
    ب
    1.12
     ৬৬
    1.09
    ल्फी
    1.09
    리의
    1.09
     kullanılan
    1.09
     実
    1.07
    ﺪ
    1.07
    Act Density 0.000%

    No Known Activations