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

    Neuronpedia

    INDEX
    Explanations

    positive adjectives, particularly the term "pretty"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    endpush
    -0.84
     насељу
    -0.71
    sidemargin
    -0.70
    InSection
    -0.69
    argout
    -0.69
    مراجع
    -0.68
     africains
    -0.68
    tanleria
    -0.67
     يتيمه
    -0.66
    Становништво
    -0.66
    POSITIVE LOGITS
     pretty
    3.21
    pretty
    2.90
    Pretty
    2.86
     Pretty
    2.75
     PRET
    2.03
    PRET
    1.39
     fairly
    1.37
     prettiest
    1.32
    retty
    1.24
     ziemlich
    1.22
    Act Density 0.094%

    No Known Activations