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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating incompleteness or uncertainty

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "completely"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    maid
    -0.91
    liest
    -0.87
    pires
    -0.73
     Emin
    -0.71
    soever
    -0.67
     Occupations
    -0.66
    mere
    -0.66
    llor
    -0.66
    pring
    -0.66
    imeters
    -0.66
    POSITIVE LOGITS
     overhaul
    0.80
     accomplished
    0.72
     reorgan
    0.71
     annihil
    0.69
     exting
    0.67
     redesign
    0.67
     reliant
    0.67
    ogn
    0.67
    âĵĺ
    0.65
     overlap
    0.65
    Act Density 0.020%

    No Known Activations