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

    Neuronpedia

    INDEX
    Explanations

    words related to quick movements or actions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the abbreviation "br," indicating a focus on specific formatting or structure within the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Decay
    -0.76
    ãĤ®
    -0.73
    EVA
    -0.73
    WIND
    -0.70
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    -0.70
    DAY
    -0.66
    WAY
    -0.64
    WARE
    -0.64
    ãĥ³ãĤ¸
    -0.64
    ãĤ»
    -0.64
    POSITIVE LOGITS
    itt
    1.00
    ighter
    0.95
    idget
    0.92
    indle
    0.92
    acing
    0.91
    atton
    0.91
    acer
    0.91
    ackets
    0.91
    imming
    0.89
    ights
    0.89
    Act Density 0.012%

    No Known Activations