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

    Neuronpedia

    INDEX
    Explanations

    time-related words and durations

    oai_token-act-pair · gpt-3.5-turbo

    references to durations of time spent on activities or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    idth
    -0.80
     Adds
    -0.75
    ãĥ¤
    -0.72
     Coverage
    -0.66
     Starts
    -0.64
    idespread
    -0.64
    rium
    -0.64
    orthy
    -0.63
    });
    -0.63
    intendent
    -0.62
    POSITIVE LOGITS
     researching
    1.19
     studying
    1.11
     cultivating
    1.10
     debating
    1.09
     staring
    1.09
     contemplating
    1.07
     preparing
    1.07
     immersed
    1.07
     dreaming
    1.06
     trying
    1.04
    Act Density 0.109%

    No Known Activations