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

    Neuronpedia

    INDEX
    Explanations

    short characters

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     مقدم
    0.64
    ார்ப
    0.63
    OrDefault
    0.62
    提
    0.60
    habt
    0.58
     Enfin
    0.58
    árez
    0.58
    ารถ
    0.58
    тую
    0.57
     नहीं
    0.57
    POSITIVE LOGITS
     i
    4.74
     ie
    3.56
    i
    2.65
     ii
    2.31
     eg
    2.07
    ie
    2.00
     e
    1.97
     অর্থাৎ
    1.94
    ｉ
    1.91
     अर्थात्
    1.80
    Act Density 0.578%

    No Known Activations