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

    Neuronpedia

    INDEX
    Explanations

    phrases related to specific formatting characters

    oai_token-act-pair · gpt-3.5-turbo

    references to specific symbols or characters that are part of a unique coding system

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     board
    -0.72
     PST
    -0.71
     Bengal
    -0.70
     PD
    -0.67
     ARC
    -0.66
     PB
    -0.63
     boards
    -0.63
     Springer
    -0.62
     Berlin
    -0.61
     GP
    -0.61
    POSITIVE LOGITS
    Ĵ
    1.40
    ª
    1.34
    IJ
    1.28
    ¹
    1.28
    ı
    1.24
    ij
    1.24
    ł
    1.16
    «
    1.15
    ®
    1.15
    §
    1.14
    Act Density 0.156%

    No Known Activations