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

    Neuronpedia

    INDEX
    Explanations

    phrases related to scientific topics or concepts

    oai_token-act-pair · gpt-3.5-turbo

    references to scientific concepts and discussions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    drops
    -0.76
    zik
    -0.70
    matched
    -0.68
    erous
    -0.67
    tower
    -0.67
    atra
    -0.66
    pora
    -0.66
    skirts
    -0.65
     capped
    -0.65
    ipop
    -0.65
    POSITIVE LOGITS
     fiction
    1.01
     curiosity
    0.97
     literacy
    0.93
     misconduct
    0.88
    ĨĴ
    0.87
     research
    0.86
     inquiry
    0.85
     scientist
    0.84
     sciences
    0.84
     science
    0.82
    Act Density 0.019%

    No Known Activations