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

    Neuronpedia

    INDEX
    Explanations

    parentheses and other punctuation

    oai_token-act-pair · gpt-3.5-turbo

    various forms of the character "âĢ"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     disse
    -0.82
     continental
    -0.78
     shroud
    -0.69
     antip
    -0.68
     ric
    -0.67
     mesmer
    -0.66
     unmarked
    -0.66
     mimic
    -0.66
     snail
    -0.66
     voic
    -0.66
    POSITIVE LOGITS
    Ŀ
    1.90
    ¡
    1.46
    ľ
    1.30
    ¦
    1.28
    Ĵ
    1.23
    ¤
    1.21
    ª
    1.21
    ĸ
    1.20
    ī
    1.19
    Ķ
    1.19
    Act Density 0.218%

    No Known Activations