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

    Neuronpedia

    INDEX
    Explanations

    references to time-related concepts, particularly focusing on the idea of occurrences happening earlier than expected

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating timeframes or deadlines

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Brav
    -0.75
     Rooms
    -0.64
     Veil
    -0.63
    UCT
    -0.63
     Cumm
    -0.63
     Butterfly
    -0.63
    qua
    -0.63
    vation
    -0.62
    watching
    -0.61
     Consent
    -0.61
    POSITIVE LOGITS
     tong
    0.73
     practicable
    0.69
    keye
    0.67
     AS
    0.67
    arn
    0.64
     demanded
    0.62
     nodd
    0.60
     compiled
    0.60
    theless
    0.59
     elaborated
    0.59
    Act Density 0.061%

    No Known Activations