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

    Neuronpedia

    INDEX
    Explanations

    comparative phrases expressing improvement or preference

    oai_token-act-pair · gpt-3.5-turbo

    phrases that express the concept of improvement or enhancement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Dresden
    -0.67
    aly
    -0.67
     Pione
    -0.64
    idon
    -0.63
    heter
    -0.63
    mberg
    -0.62
    anuts
    -0.61
    ategory
    -0.61
     Canaver
    -0.61
    leted
    -0.60
    POSITIVE LOGITS
     than
    1.17
     suited
    1.06
     behaved
    0.95
    seller
    0.90
     Than
    0.84
    than
    0.82
     acquainted
    0.80
     quality
    0.78
     vers
    0.75
    iating
    0.75
    Act Density 0.039%

    No Known Activations