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

    Neuronpedia

    INDEX
    Explanations

    references to different types of official documents or content citations

    oai_token-act-pair · gpt-3.5-turbo

    instances of content in brackets or brackets themselves, indicating cited content or references

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     hog
    -0.81
     seams
    -0.74
     lemon
    -0.72
     consumption
    -0.69
     uncomp
    -0.68
     clash
    -0.67
     uncertain
    -0.67
     intensive
    -0.66
     emit
    -0.66
     imperson
    -0.65
    POSITIVE LOGITS
    Pg
    1.35
    âĢ¦]
    1.32
    ...]
    1.27
    sic
    1.06
    interstitial
    1.05
    src
    1.04
    paragraph
    1.02
    nb
    1.00
    Footnote
    0.99
    ?]
    0.93
    Act Density 0.017%

    No Known Activations