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

    Neuronpedia

    INDEX
    Explanations

    various characters, symbols, and words in a different language, possibly as part of a specific format or encoding

    oai_token-act-pair · gpt-3.5-turbo

    characters and symbols from non-Latin alphabets, particularly Cyrillic or similar scripts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    bett
    -0.75
     reconc
    -0.74
    iqueness
    -0.73
    urion
    -0.73
    irtual
    -0.69
    drawn
    -0.69
    merce
    -0.68
     crossover
    -0.68
    clud
    -0.68
    nown
    -0.67
    POSITIVE LOGITS
     Ð
    1.17
    Ø©
    1.17
     ÙĪ
    1.15
    н
    1.15
    и
    1.12
    Ñı
    1.10
     è£ıç
    1.08
    е
    1.08
    1.08
    ÏĤ
    1.06
    Act Density 0.049%

    No Known Activations