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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating comparison or addition, with an emphasis on the word "as"

    oai_token-act-pair · gpt-3.5-turbo

    the word "as" used in various contexts throughout the document

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ager
    -0.80
    eks
    -0.80
    erb
    -0.76
    reply
    -0.74
    iddle
    -0.74
    oder
    -0.74
    god
    -0.74
    rones
    -0.73
    anos
    -0.72
    amon
    -0.72
    POSITIVE LOGITS
     ensuring
    0.85
     assorted
    0.84
     those
    0.84
     allowing
    0.79
     providing
    0.78
     being
    0.78
     occasional
    0.77
     other
    0.74
     some
    0.74
     optionally
    0.74
    Act Density 0.072%

    No Known Activations