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

    Neuronpedia

    INDEX
    Explanations

    programming keywords and separators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     negras
    0.41
     छोटे
    0.40
     χώ
    0.40
     गाड़ी
    0.39
    িকপ্ট
    0.39
     большой
    0.39
     أر
    0.39
     כן
    0.39
    becca
    0.38
     νο
    0.38
    POSITIVE LOGITS
    s
    0.60
    н
    0.54
    ב
    0.53
    and
    0.52
    i
    0.51
    ه
    0.50
    ה
    0.48
    ন
    0.47
    x
    0.47
    न
    0.46
    Act Density 0.190%

    No Known Activations