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

    Neuronpedia

    INDEX
    Explanations

    titles of sections in a piece of text

    oai_token-act-pair · gpt-3.5-turbo

    references to editing or sections within a document

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    milo
    -0.69
    plet
    -0.66
    gren
    -0.65
     manif
    -0.62
    ocket
    -0.61
     incarcer
    -0.61
    mable
    -0.60
    kinson
    -0.60
     jeans
    -0.58
    SourceFile
    -0.58
    POSITIVE LOGITS
     edit
    0.78
    onal
    0.70
     Blizzard
    0.69
     ][
    0.67
     ]
    0.66
     Franç
    0.65
     âĨij
    0.64
     Edit
    0.64
    âĶĢâĶĢ
    0.64
     Torrent
    0.62
    Act Density 0.022%

    No Known Activations