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

    Neuronpedia

    INDEX
    Explanations

    comparisons of quantities, specifically indicating a value as being multiple times higher than another

    oai_token-act-pair · gpt-3.5-turbo

    references to comparative ratios or multiplicative differences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Reviewer
    -0.99
    ITAL
    -0.77
    irt
    -0.77
    ogy
    -0.74
    aceous
    -0.73
    ItemTracker
    -0.73
    uld
    -0.71
    tymology
    -0.71
    riad
    -0.71
    liction
    -0.71
    POSITIVE LOGITS
    cale
    1.12
    paces
    0.87
    bp
    0.78
    hops
    0.76
    bps
    0.70
    fold
    0.70
    cus
    0.69
    pan
    0.68
     consuming
    0.68
    borough
    0.67
    Act Density 0.024%

    No Known Activations