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

    Neuronpedia

    INDEX
    Explanations

    numeric quantities related to the length of written text

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of page numbers or references to document length

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    oÄŁ
    -0.86
    ereo
    -0.77
     Yel
    -0.77
    ratulations
    -0.70
    nder
    -0.69
    igham
    -0.68
    FML
    -0.66
    ignty
    -0.65
    eworld
    -0.65
    ivia
    -0.64
    POSITIVE LOGITS
     detailing
    1.01
     pages
    0.98
     outlining
    0.96
     detail
    0.89
     documents
    0.89
     describing
    0.86
     memos
    0.86
     manuscript
    0.85
     document
    0.84
     outlines
    0.81
    Act Density 0.036%

    No Known Activations