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

    Neuronpedia

    INDEX
    Explanations

    specific positional or contextual references in a text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    ranking, position, order

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    رشف
    -0.54
    parseLong
    -0.53
     يتيمه
    -0.51
    carbo
    -0.51
    commercial
    -0.50
     scand
    -0.50
    AddTagHelper
    -0.50
    hydro
    -0.50
     San
    -0.49
     Lark
    -0.49
    POSITIVE LOGITS
    位
    2.41
     位
    1.83
    位的
    1.50
    位の
    1.34
    위
    1.11
    一位
    1.02
     vị
    0.93
     POSITION
    0.81
     Position
    0.81
    三位
    0.80
    Act Density 0.001%

    No Known Activations