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

    Neuronpedia

    INDEX
    Explanations

    phrases containing hyphens associated with intensity or emphasis

    oai_token-act-pair · gpt-3.5-turbo

    patterns related to full-length durations or completions in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Tycoon
    -0.72
    Ń·
    -0.68
     Mystic
    -0.68
     Pok
    -0.66
     preced
    -0.65
     Lum
    -0.61
     Kai
    -0.59
     Super
    -0.59
     chants
    -0.59
     Pes
    -0.59
    POSITIVE LOGITS
    fledged
    1.72
    length
    1.54
    blown
    1.50
    scale
    1.45
    blooded
    1.40
    thro
    1.37
    bodied
    1.37
    sized
    1.35
    contact
    1.29
    strength
    1.27
    Act Density 0.022%

    No Known Activations