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

    Neuronpedia

    INDEX
    Explanations

    the word "not" with emphasis

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing negation or rejection

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    å¥
    -0.81
    cano
    -0.72
    çļ
    -0.70
    alez
    -0.69
    kamp
    -0.67
    ously
    -0.67
    ounters
    -0.67
     peaks
    -0.66
    iox
    -0.65
    ãĤ¼
    -0.65
    POSITIVE LOGITS
     gonna
    1.21
     afraid
    1.16
     kidding
    1.15
     interested
    1.14
     bothered
    1.14
     necessarily
    1.12
     ashamed
    1.07
     aware
    1.04
    epad
    1.02
     amused
    1.02
    Act Density 0.112%

    No Known Activations