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

    Neuronpedia

    INDEX
    Explanations

    first-person references combined with personal opinions and observations

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate personal opinions or experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    eland
    -0.67
    Dispatch
    -0.67
     Emin
    -0.65
     Definitions
    -0.65
     Stall
    -0.64
    LY
    -0.63
     Militia
    -0.62
    ENCY
    -0.62
     Sparkle
    -0.62
     Continuous
    -0.62
    POSITIVE LOGITS
     identifiable
    1.02
     invested
    0.83
     benefited
    0.83
     intervened
    0.81
     opposed
    0.79
    ised
    0.79
     benef
    0.78
    ographed
    0.77
     attest
    0.77
     thanked
    0.76
    Act Density 0.026%

    No Known Activations