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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating someone or something located behind a person or object

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "behind."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    istic
    -0.77
     Pwr
    -0.73
    ogl
    -0.67
    acci
    -0.65
    iser
    -0.64
    olid
    -0.63
    ISM
    -0.62
    cki
    -0.61
    ucc
    -0.60
    ander
    -0.60
    POSITIVE LOGITS
    âĸ¬âĸ¬
    0.79
    shore
    0.78
    ¬
    0.70
    coat
    0.70
    paren
    0.70
    */(
    0.69
    sight
    0.67
     bars
    0.66
     behind
    0.66
    gression
    0.66
    Act Density 0.019%

    No Known Activations