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

    Neuronpedia

    INDEX
    Explanations

    sequences involving a specific symbol, represented in the activations by 'âĢ'

    oai_token-act-pair · gpt-3.5-turbo

    the character representation of a symbol or emoticon

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Tunis
    -0.77
     Libyan
    -0.75
     Kenyan
    -0.73
     scattering
    -0.69
     guided
    -0.69
     diffusion
    -0.67
     Eisen
    -0.64
     guidance
    -0.64
     Counsel
    -0.63
     memorandum
    -0.63
    POSITIVE LOGITS
    ¬
    1.31
    ¡
    1.27
    ¹
    1.27
    ½
    1.23
    «
    1.21
    Ń
    1.21
    ¿
    1.21
    į
    1.19
    Į
    1.19
    ª
    1.18
    Act Density 0.327%

    No Known Activations