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

    Neuronpedia

    INDEX
    Explanations

    the superlative form of adjectives

    oai_token-act-pair · gpt-3.5-turbo

    the word "most" and related phrases indicating superiority or extremity in context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    pload
    -0.75
    rompt
    -0.66
    ategor
    -0.65
    instead
    -0.64
     Travels
    -0.62
     Handling
    -0.60
     Oaks
    -0.59
    æ©
    -0.59
    cia
    -0.59
    ences
    -0.59
    POSITIVE LOGITS
     important
    0.93
     notable
    0.87
     profitable
    0.86
     importantly
    0.85
     obvious
    0.84
    likely
    0.84
     noticeable
    0.83
    iest
    0.82
     recognizable
    0.80
     prominent
    0.80
    Act Density 0.076%

    No Known Activations