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

    Neuronpedia

    INDEX
    Explanations

    . or ] followed by code keywords

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    k
    1.16
     to
    1.02
     on
    0.93
     as
    0.93
    in
    0.92
     and
    0.91
    on
    0.90
     an
    0.87
    en
    0.85
     in
    0.85
    POSITIVE LOGITS
    ﺖ
    0.79
     приез
    0.79
    ี
    0.77
     elegido
    0.76
    ONES
    0.75
     ascenso
    0.75
     premio
    0.75
     ﺍﻟ
    0.73
     piedi
    0.73
     visto
    0.72
    Act Density 0.196%

    No Known Activations