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

    Neuronpedia

    INDEX
    Explanations

    avoiding direct data exchange

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    滏
    0.51
     দেখা
    0.49
    矨
    0.48
    酤
    0.48
    姮
    0.47
    odhya
    0.47
    હેર
    0.46
     ശബരിമല
    0.46
    睽
    0.45
    䐍
    0.45
    POSITIVE LOGITS
    ga
    0.48
    ↵↵
    0.45
    lio
    0.44
    kor
    0.43
    /
    0.43
    1
    0.42
    hz
    0.40
     pleno
    0.40
    cm
    0.39
     serialized
    0.39
    Act Density 0.002%

    No Known Activations