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

    Neuronpedia

    INDEX
    Explanations

    words related to tangled, jumbled, or mixed-up text

    oai_token-act-pair · gpt-3.5-turbo

    patterns of repeating syllables or phonetic sounds

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     suspic
    -0.68
     convol
    -0.68
     combust
    -0.66
     SERV
    -0.65
     reluct
    -0.64
    Adds
    -0.64
     sweats
    -0.62
     hazard
    -0.60
     scrape
    -0.60
    FIX
    -0.60
    POSITIVE LOGITS
    ensis
    1.05
    ocious
    0.74
    aii
    0.73
     ('
    0.72
    [_
    0.71
    çļĦ
    0.71
    )].
    0.67
    wyn
    0.66
    iii
    0.66
    ival
    0.66
    Act Density 0.358%

    No Known Activations