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

    Neuronpedia

    INDEX
    Explanations

    adjectives ending in 'y'

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    uther
    -0.68
    channelAvailability
    -0.61
    arthed
    -0.61
    roit
    -0.60
    RF
    -0.60
    oyer
    -0.57
    ovember
    -0.57
    arte
    -0.56
    anamo
    -0.55
    ainer
    -0.55
    POSITIVE LOGITS
    ness
    0.81
     eyed
    0.78
    glers
    0.70
    nesses
    0.68
     enough
    0.66
    ly
    0.64
     little
    0.61
     sounding
    0.60
     tendencies
    0.58
     seas
    0.57
    Act Density 12.788%

    No Known Activations