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

    Neuronpedia

    INDEX
    Explanations

    numerical values preceded by a specific number

    oai_token-act-pair · gpt-3.5-turbo

    specific numeric values, particularly "52" and "53", often in a statistical or percentage context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    urus
    -0.67
    encers
    -0.66
    phrine
    -0.65
    spread
    -0.65
     Scouting
    -0.63
     streamed
    -0.62
    razil
    -0.62
     repr
    -0.61
     Parade
    -0.60
    gain
    -0.60
    POSITIVE LOGITS
    ielding
    0.89
    ield
    0.87
    00
    0.84
    andowski
    0.73
    itely
    0.72
    50
    0.71
    IFE
    0.71
    oji
    0.69
    yo
    0.69
    ramer
    0.68
    Act Density 0.029%

    No Known Activations