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

    Neuronpedia

    INDEX
    Explanations

    references to surfing or skateboarding

    oai_token-act-pair · gpt-3.5-turbo

    references to waterboarding and related activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    aments
    -0.75
    ively
    -0.69
     Bounty
    -0.67
     Frey
    -0.66
    talk
    -0.65
    ures
    -0.65
    edy
    -0.65
    mates
    -0.63
    orf
    -0.63
     Toll
    -0.62
    POSITIVE LOGITS
    yip
    0.91
    anchester
    0.87
     surfing
    0.85
    ancock
    0.84
    boarding
    0.80
     biking
    0.73
    xtap
    0.72
     referen
    0.71
    quished
    0.71
    ailand
    0.70
    Act Density 0.025%

    No Known Activations