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

    Neuronpedia

    INDEX
    Explanations

    company

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ഹിന്ദു
    0.39
     ٣
    0.38
     sít
    0.38
    。「
    0.37
     are
    0.36
    ٦
    0.36
    𝐥
    0.36
     يَ
    0.36
     香港
    0.36
     parâ
    0.35
    POSITIVE LOGITS
    t
    0.57
    ir
    0.56
    ار
    0.51
    ing
    0.51
    ’
    0.51
    g
    0.49
    v
    0.48
    ad
    0.48
    X
    0.46
    ik
    0.44
    Act Density 0.216%

    No Known Activations