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

    Neuronpedia

    INDEX
    Explanations

    phrases with the word "Mill" followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of the word "Mill."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     fuse
    -0.63
     compelling
    -0.63
     Ae
    -0.62
     Bahamas
    -0.62
     redress
    -0.61
     squash
    -0.60
    /-
    -0.59
    nces
    -0.59
    LY
    -0.59
     Chak
    -0.58
    POSITIVE LOGITS
    enium
    1.74
    isec
    1.41
    icent
    1.40
    imeter
    1.40
    ennial
    1.38
    imet
    1.37
    imeters
    1.33
    inery
    1.30
    iband
    1.30
    enn
    1.24
    Act Density 0.034%

    No Known Activations