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

    Neuronpedia

    INDEX
    Explanations

    sequences of underscore characters

    oai_token-act-pair · gpt-3.5-turbo

    patterns indicating visual elements or formatting in a document

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ollar
    -0.68
    pora
    -0.65
    lling
    -0.62
    anan
    -0.62
    ãĥ´
    -0.62
    arest
    -0.61
     clipping
    -0.60
     denying
    -0.59
     Gunn
    -0.58
     mull
    -0.58
    POSITIVE LOGITS
    */
    0.99
    *****
    0.93
     =================================================================
    0.85
    edit
    0.78
    Requirements
    0.77
    ***
    0.76
    ---
    0.76
    ===
    0.74
    quote
    0.74
    ------
    0.72
    Act Density 0.045%

    No Known Activations