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

    Neuronpedia

    INDEX
    Explanations

    words or phrases related to providing information or completing a task

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to the concept of "filling in" or completing information

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    sed
    -0.61
    pi
    -0.61
    nos
    -0.61
    iral
    -0.61
    xual
    -0.60
     Fiesta
    -0.60
     Feast
    -0.59
    velt
    -0.59
    ials
    -0.58
    cia
    -0.57
    POSITIVE LOGITS
     gaps
    1.15
     loopholes
    0.78
     holes
    0.75
     gap
    0.73
     inconsistencies
    0.73
     details
    0.73
    missing
    0.70
     dots
    0.68
     sten
    0.66
     blank
    0.66
    Act Density 0.069%

    No Known Activations