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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "hug"

    oai_token-act-pair · gpt-3.5-turbo

    references to hugging

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    iasis
    -0.72
    piracy
    -0.67
     piracy
    -0.65
     Punk
    -0.65
    DoS
    -0.62
    ourses
    -0.61
    protected
    -0.60
     dial
    -0.59
    secondary
    -0.58
     Pir
    -0.58
    POSITIVE LOGITS
    glers
    1.03
     goodbye
    1.03
    gers
    1.02
    eness
    1.02
     Hug
    0.98
     hug
    0.96
     hugs
    0.96
    aneers
    0.94
    wrap
    0.90
    eson
    0.88
    Act Density 0.013%

    No Known Activations