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

    Neuronpedia

    INDEX
    Explanations

    C format specifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ීම්
    0.45
    ඇ
    0.44
    数字
    0.41
    split
    0.41
    हिट
    0.41
    Pro
    0.40
     թվ
    0.40
    unic
    0.39
    CAL
    0.38
     paralelas
    0.37
    POSITIVE LOGITS
     conversions
    0.49
     Conversion
    0.48
     spec
    0.46
     Spec
    0.44
     conversion
    0.44
     специ
    0.42
     escapes
    0.40
    Conversion
    0.40
     escaping
    0.39
     escaped
    0.38
    Act Density 0.007%

    No Known Activations