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

    Neuronpedia

    INDEX
    Explanations

    instances where an action of pulling or drawing something is described

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "pulled."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ibel
    -0.78
    voy
    -0.78
    unic
    -0.74
    esan
    -0.74
    icol
    -0.72
    brates
    -0.69
    esa
    -0.69
    ggies
    -0.68
    elsius
    -0.68
    heid
    -0.68
    POSITIVE LOGITS
    pull
    0.99
     pull
    0.92
     aggro
    0.92
    Pull
    0.87
     levers
    0.80
     out
    0.74
     weeds
    0.73
    chairs
    0.73
     pulling
    0.72
    quote
    0.72
    Act Density 0.013%

    No Known Activations