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

    Neuronpedia

    INDEX
    Explanations

    tons, far, adjacent, huge, everything

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    take
    1.03
    evil
    1.00
    ی
    0.98
    hard
    0.95
    eat
    0.95
     uplifting
    0.92
     fateful
    0.90
    come
    0.89
    i
    0.89
    e
    0.89
    POSITIVE LOGITS
     ಇತರ
    0.93
    ிருக்கு
    0.86
     exd
    0.85
    wString
    0.82
     Хабаровского
    0.82
    ικοί
    0.81
     στους
    0.80
     επίσης
    0.80
    ויות
    0.79
     очень
    0.77
    Act Density 0.245%

    No Known Activations