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

    Neuronpedia

    INDEX
    Explanations

    numerical indicators such as percentages or quantities

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing scarcity or low quantity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    agher
    -0.66
     weaving
    -0.63
    obo
    -0.62
    NW
    -0.62
    Draft
    -0.60
     Bust
    -0.59
     widening
    -0.59
    dress
    -0.58
    irst
    -0.58
    alam
    -0.58
    POSITIVE LOGITS
     hundred
    0.86
    een
    0.84
     dozen
    0.80
     mortals
    0.77
    est
    0.76
    ones
    0.74
    eenth
    0.73
    orem
    0.73
    ishers
    0.71
     thousand
    0.71
    Act Density 0.026%

    No Known Activations