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

    Neuronpedia

    INDEX
    Explanations

    possessive pronouns followed by positive adjectives

    oai_token-act-pair · gpt-3.5-turbo

    possessive pronouns and expressions of personal opinion or preference

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    boro
    -0.75
    ariat
    -0.75
    amia
    -0.74
    igslist
    -0.73
    ifice
    -0.72
     existed
    -0.72
    etz
    -0.71
    ĸļ
    -0.71
    erver
    -0.71
    ecast
    -0.70
    POSITIVE LOGITS
     thoughts
    1.57
     favorite
    1.54
     favourite
    1.51
     favorites
    1.39
     impressions
    1.32
     favourites
    1.26
     opinion
    1.23
     Thoughts
    1.16
    favorite
    1.14
     experiences
    1.13
    Act Density 0.157%

    No Known Activations