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

    Neuronpedia

    INDEX
    Explanations

    elements related to character encoding and special symbols

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    zem
    -0.17
    cgi
    -0.16
    cott
    -0.16
    ustum
    -0.15
    kest
    -0.15
     Clarkson
    -0.14
    andy
    -0.14
    лÑıн
    -0.14
    formats
    -0.13
     Engineering
    -0.13
    POSITIVE LOGITS
     characters
    0.44
     character
    0.43
     letters
    0.41
     char
    0.38
     Characters
    0.37
     Character
    0.36
     letter
    0.35
     Letters
    0.35
    -char
    0.35
     chars
    0.35
    Act Density 0.165%

    No Known Activations