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

    Neuronpedia

    INDEX
    Explanations

    overmixing develops gluten

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     exces
    0.63
     reminis
    0.60
    ழுப்பு
    0.59
    两大
    0.58
     insuf
    0.58
    天津
    0.57
    اہ
    0.57
    不堪
    0.56
    印度
    0.56
     Kakkar
    0.56
    POSITIVE LOGITS
    t
    0.96
    d
    0.90
    c
    0.78
    1
    0.74
    ר
    0.73
     for
    0.68
    ↵↵↵
    0.67
    л
    0.66
    ק
    0.65
    r
    0.64
    Act Density 0.008%

    No Known Activations