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

    Neuronpedia

    INDEX
    Explanations

    phrases enclosed in quotation marks indicating a statement or dialogue

    oai_token-act-pair · gpt-3.5-turbo

    sentences that contain quotes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     traveling
    -0.55
     halftime
    -0.53
     gray
    -0.53
     neighbors
    -0.52
     POLITICO
    -0.50
     labor
    -0.50
     nighttime
    -0.50
     longtime
    -0.49
    ĪĴ
    -0.49
     alum
    -0.48
    POSITIVE LOGITS
    ".
    3.35
    !".
    3.07
    ".[
    2.99
    ?".
    2.92
    ''.
    2.59
    ",
    2.49
    '."
    2.45
    "!
    2.43
    ").
    2.39
     ".
    2.32
    Act Density 0.019%

    No Known Activations