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

    Neuronpedia

    INDEX
    Explanations

    latex formatting and citations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Архив
    0.77
    )²
    0.76
     ½
    0.75
    inguishable
    0.74
    cula
    0.74
     Yar
    0.74
    हारिक
    0.74
    砾
    0.74
    foresaid
    0.74
    ysuckle
    0.73
    POSITIVE LOGITS
    emph
    1.75
    textit
    1.39
    textbf
    1.21
     emph
    0.96
    underline
    0.96
    (\
    0.93
    texttt
    0.92
    verb
    0.92
    gl
    0.90
    emphasis
    0.89
    Act Density 0.002%

    No Known Activations