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

    Neuronpedia

    INDEX
    Explanations

    words related to coding, programming, and scripting

    oai_token-act-pair · gpt-3.5-turbo

    terms related to scripted and coded content or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ingham
    -0.83
    abet
    -0.82
    alos
    -0.76
    our
    -0.76
    Scot
    -0.74
    allows
    -0.73
    went
    -0.71
    adder
    -0.70
    fell
    -0.70
    aders
    -0.69
    POSITIVE LOGITS
    uled
    0.86
    irect
    0.78
    ãĥĥãĤ¯
    0.75
    ãĥ¼ãĤ¯
    0.75
    nesday
    0.74
    ĸļ
    0.73
     unintention
    0.70
    urally
    0.68
    ocument
    0.68
    escription
    0.67
    Act Density 0.057%

    No Known Activations