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

    Neuronpedia

    INDEX
    Explanations

    numbers and quantities

    oai_token-act-pair · gpt-3.5-turbo

    cardinal numbers (two, three, four, eight, ten, etc.) at the beginning of phrases or sentences.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @aserino3

    numerical values, especially those indicating quantities or measurements.

    oai_token-act-pair · gpt-4oTriggered by @459865181

    numerical data and statistics related to counts or quantities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    geist
    -0.67
    AMI
    -0.63
    othal
    -0.60
    ovan
    -0.59
     Kun
    -0.56
     totality
    -0.55
    iatus
    -0.55
    ablishment
    -0.54
    normal
    -0.54
     jo
    -0.54
    POSITIVE LOGITS
    etheless
    0.82
    thirds
    0.74
     thirds
    0.71
    doms
    0.70
     times
    0.70
    tions
    0.65
     fold
    0.63
     Weir
    0.62
    taboola
    0.62
    pecially
    0.61
    Act Density 0.136%

    No Known Activations