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

    Neuronpedia

    INDEX
    Explanations

    language ending in です, ます, 있다, or da

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    স্পতি
    0.51
    舴
    0.50
    跚
    0.49
    𝙖
    0.48
    㓡
    0.48
    𝙝
    0.48
    敺
    0.48
    धित
    0.47
    ल्लाला
    0.47
    𝙨
    0.47
    POSITIVE LOGITS
    。
    0.52
    ↵
    0.49
    です
    0.49
     것이다
    0.48
     제
    0.48
    다
    0.47
     ซึ่ง
    0.47
    な
    0.47
    .
    0.46
     그
    0.46
    Act Density 0.006%

    No Known Activations