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

    Neuronpedia

    INDEX
    Explanations

    words related to leaving or exit

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "left"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    MAT
    -0.75
    externalActionCode
    -0.71
    ITNESS
    -0.68
    oun
    -0.68
    Temperature
    -0.66
    MpServer
    -0.65
    CONT
    -0.64
    ount
    -0.63
    OTAL
    -0.63
    inarily
    -0.62
    POSITIVE LOGITS
    wing
    0.95
    overs
    0.94
     wing
    0.83
     left
    0.80
     handed
    0.72
    left
    0.71
     Left
    0.71
     undone
    0.71
    uve
    0.70
    ward
    0.69
    Act Density 0.033%

    No Known Activations