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

    Neuronpedia

    INDEX
    Explanations

    a perfect

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ید
    0.90
    بی
    0.81
    1
    0.76
    }\
    0.73
    고
    0.68
     heart
    0.68
    し
    0.66
     lukewarm
    0.66
    산
    0.66
     heartbreaking
    0.65
    POSITIVE LOGITS
    k
    1.08
    า
    0.99
    as
    0.96
    0.88
    er
    0.86
    ா
    0.82
    supportsFocus
    0.76
    ا
    0.76
    an
    0.75
    ా
    0.75
    Act Density 0.000%

    No Known Activations