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

    Neuronpedia

    INDEX
    Explanations

    saved copy and model

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    '
    0.92
    গ
    0.89
    ي
    0.88
    i
    0.85
    на
    0.83
    م
    0.80
    도
    0.79
    の為
    0.79
    も
    0.77
    시
    0.76
    POSITIVE LOGITS
     Problems
    0.84
     Assessment
    0.77
     =
    0.74
    тно
    0.73
     an
    0.72
     ={
    0.70
     on
    0.69
    ervice
    0.69
     सर्विसेज
    0.68
    с
    0.66
    Act Density 0.000%

    No Known Activations