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

    Neuronpedia

    INDEX
    Explanations

    URLs with numbers and `com`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     বেশিরভাগ
    0.18
     বেশকিছু
    0.18
     Muslim
    0.18
    ähn
    0.18
    ığınız
    0.17
    īga
    0.17
    Algun
    0.17
     Buddhist
    0.17
     स्वास्थ्य
    0.17
     Besides
    0.17
    POSITIVE LOGITS
    那个
    0.20
     没有
    0.20
     breezes
    0.19
     mattered
    0.18
    这个时候
    0.18
    依旧
    0.18
     rumble
    0.17
     that
    0.17
     ấy
    0.17
    看上去
    0.17
    Act Density 0.002%

    No Known Activations