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

    Neuronpedia

    INDEX
    Explanations

    words related to leaving or exiting

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "departure."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    verbs
    -0.89
    ollo
    -0.78
    otin
    -0.76
    essee
    -0.75
    kef
    -0.74
    acca
    -0.74
    aughtered
    -0.69
    ibaba
    -0.68
    Cola
    -0.67
    Zip
    -0.67
    POSITIVE LOGITS
     departure
    0.99
     departures
    0.84
    dylib
    0.68
     depart
    0.66
    rance
    0.65
     Leaves
    0.65
     anytime
    0.64
     Hogwarts
    0.64
     altogether
    0.61
     undone
    0.60
    Act Density 0.015%

    No Known Activations