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

    Neuronpedia

    INDEX
    Explanations

    initials of topics

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Nex
    0.65
    Tec
    0.60
    Male
    0.58
    Mex
    0.54
    Spain
    0.53
    Regex
    0.53
    tipo
    0.53
    Ridge
    0.53
    M
    0.52
    Span
    0.52
    POSITIVE LOGITS
     Ironically
    0.57
     간
    0.55
     Ensuring
    0.54
     ensuring
    0.54
     Perhaps
    0.50
     가장
    0.50
     sacrificing
    0.49
     crucial
    0.48
     Thanksgiving
    0.46
    த்தார்
    0.45
    Act Density 0.130%

    No Known Activations