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

    Neuronpedia

    INDEX
    Explanations

    instances of special characters and uppercase letter sequences in a text

    oai_token-act-pair · gpt-3.5-turbo

    punctuation and formatting elements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     centrally
    -0.84
     este
    -0.82
    etheless
    -0.81
     estranged
    -0.76
     central
    -0.69
     hig
    -0.68
     fortune
    -0.68
     honoured
    -0.67
     hon
    -0.67
     anecd
    -0.65
    POSITIVE LOGITS
    Pause
    0.97
    Narr
    0.92
    Said
    0.83
    Reply
    0.82
    Everybody
    0.81
    Discussion
    0.80
    —-
    0.79
    Howard
    0.77
    Accessory
    0.77
    Instead
    0.76
    Act Density 0.184%

    No Known Activations