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

    Neuronpedia

    INDEX
    Explanations

    hyphenated words followed by numerical values, such as "best-selling" or "third-highest"

    oai_token-act-pair · gpt-3.5-turbo

    the superlative form of adjectives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Hicks
    -0.70
     Bullets
    -0.69
     Indigo
    -0.67
     sep
    -0.66
     Cox
    -0.66
     dialog
    -0.65
     Ratio
    -0.65
     Debor
    -0.65
     Daylight
    -0.64
     Kits
    -0.63
    POSITIVE LOGITS
    ever
    1.40
    ranked
    1.37
    selling
    1.28
    performing
    1.24
    worst
    1.23
    known
    1.21
    favorite
    1.19
    ranking
    1.18
    rated
    1.15
    looking
    1.10
    Act Density 0.053%

    No Known Activations