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

    Neuronpedia

    INDEX
    Explanations

    references to the word "ask"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "ask."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     cue
    -0.67
     customary
    -0.65
     nerv
    -0.65
    coni
    -0.62
     ejac
    -0.61
    BER
    -0.59
     intens
    -0.59
    bably
    -0.57
     subsid
    -0.57
    nerg
    -0.57
    POSITIVE LOGITS
    atchewan
    1.25
    ansas
    1.11
    oln
    1.02
    edIn
    1.01
     Rapids
    0.97
    ask
    0.92
    ings
    0.91
    atoon
    0.91
    aign
    0.89
    ets
    0.88
    Act Density 0.014%

    No Known Activations