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

    Neuronpedia

    INDEX
    Explanations

    phrases containing a series of hyphens and words separated by underscores

    oai_token-act-pair · gpt-3.5-turbo

    underscore characters in specific contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Heavenly
    -0.83
     Bone
    -0.81
     Thrones
    -0.80
     buoy
    -0.79
     Woody
    -0.76
     fields
    -0.76
     Yen
    -0.76
     Xiao
    -0.76
     wildly
    -0.75
     turnover
    -0.72
    POSITIVE LOGITS
    edition
    1.75
    guide
    1.63
    ebook
    1.55
    united
    1.50
    collection
    1.48
    reve
    1.47
    report
    1.46
    vs
    1.45
    conference
    1.45
    euro
    1.45
    Act Density 0.057%

    No Known Activations