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

    Neuronpedia

    INDEX
    Explanations

    dates in a specific format - month, day, and year

    oai_token-act-pair · gpt-3.5-turbo

    instances of the letter "O" in varying contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãĤ¼ãĤ¦ãĤ¹
    -0.84
     wip
    -0.83
    ãĥ¼ãĥĨãĤ£
    -0.82
    yards
    -0.74
     à¨
    -0.74
    ãĥķ
    -0.72
    ÑĮ
    -0.71
     sinks
    -0.70
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    -0.69
    ãĥ¼ãĥĨ
    -0.65
    POSITIVE LOGITS
    vernight
    1.14
    culus
    1.12
    tto
    1.11
    ceans
    1.10
    lymp
    1.10
    zzy
    1.08
    scill
    1.06
    oga
    1.06
    atmeal
    1.04
    oops
    1.00
    Act Density 0.020%

    No Known Activations