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

    Neuronpedia

    INDEX
    Explanations

    influence on or of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ्स
    3.37
    з
    2.97
     аз
    2.85
    to
    2.81
    ят
    2.77
    textit
    2.71
    ली
    2.70
    tion
    2.67
    ्तिक
    2.66
    اً
    2.66
    POSITIVE LOGITS
    en
    3.02
    о
    2.83
     Tand
    2.63
    ined
    2.60
     деву
    2.51
    儡
    2.48
    藉
    2.43
    ধ
    2.42
     crosstalk
    2.40
    clide
    2.36
    Act Density 0.026%

    No Known Activations