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

    Neuronpedia

    INDEX
    Explanations

    strings in technical format with hyphens and dates

    oai_token-act-pair · gpt-3.5-turbo

    repeated formatting or section indicators in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    igation
    -0.73
    fish
    -0.72
     repl
    -0.66
    pocket
    -0.65
    gi
    -0.64
    apolis
    -0.63
     herb
    -0.63
     observ
    -0.62
    ²¾
    -0.62
    reen
    -0.62
    POSITIVE LOGITS
     ---
    0.92
    -------
    0.91
    ---
    0.86
     ---------
    0.85
    PsyNetMessage
    0.83
    ------
    0.82
    ————
    0.80
    =-=-
    0.78
    ---------
    0.78
    enhagen
    0.77
    Act Density 0.011%

    No Known Activations