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

    Neuronpedia

    INDEX
    Explanations

    numerical values preceded by a space and followed by a period

    oai_token-act-pair · gpt-3.5-turbo

    content that poses questions or expresses curiosity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    wagen
    -0.81
    creen
    -0.77
     Mellon
    -0.73
     Canaver
    -0.71
    oulos
    -0.71
     nudity
    -0.69
     braces
    -0.68
     delegation
    -0.66
     Asheville
    -0.65
     potatoes
    -0.64
    POSITIVE LOGITS
    ________________________________________________________________
    1.06
    Ë
    0.96
    Ëľ
    0.96
    É
    0.94
    urn
    0.93
    _____
    0.93
    ________
    0.93
    Ì
    0.92
    cial
    0.92
    ~~~~~~~~~~~~~~~~
    0.90
    Act Density 0.006%

    No Known Activations