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

    Neuronpedia

    INDEX
    Explanations

    hyphenated numerical ranges that end in a specific digit

    oai_token-act-pair · gpt-3.5-turbo

    references to one-off events or concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Hoo
    -0.71
     SIG
    -0.66
    ITED
    -0.65
    ãĤ¼ãĤ¦ãĤ¹
    -0.65
    emetery
    -0.64
     Dag
    -0.63
    ynski
    -0.63
    ÃįÃį
    -0.61
     vel
    -0.60
     wedd
    -0.60
    POSITIVE LOGITS
    sided
    1.32
    minute
    1.20
    year
    1.19
    dimensional
    1.19
    handed
    1.11
    tailed
    1.09
    piece
    1.08
    day
    1.08
    legged
    1.07
    month
    1.05
    Act Density 0.041%

    No Known Activations