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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating personal opinions or preferences

    oai_token-act-pair · gpt-3.5-turbo

    emphasized statements and expressions of personal opinion

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    »Ĵ
    -0.89
    WIND
    -0.73
    interstitial
    -0.73
    æĸ
    -0.69
    IRT
    -0.69
    çĭ
    -0.68
    ĸļ
    -0.68
    ANC
    -0.66
    ļéĨĴ
    -0.66
    ãĥĺ
    -0.66
    POSITIVE LOGITS
    but
    1.30
     But
    1.30
     but
    1.29
    But
    1.09
     BUT
    1.09
     However
    1.07
    though
    0.96
    However
    0.93
    although
    0.92
     Nonetheless
    0.89
    Act Density 0.757%

    No Known Activations