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

    Neuronpedia

    INDEX
    Explanations

    endings of sentences

    oai_token-act-pair · gpt-3.5-turbo

    positive statements or conclusions related to societal or communal issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     rou
    -0.73
     poked
    -0.72
     hatch
    -0.72
     delinquent
    -0.72
    lled
    -0.72
    raph
    -0.70
     unlucky
    -0.70
     himself
    -0.69
     mur
    -0.69
     inflated
    -0.68
    POSITIVE LOGITS
     Ultimately
    1.18
     Therefore
    1.15
     Thankfully
    1.08
     Flavoring
    1.06
     Such
    1.06
     Serving
    1.05
     Unfortunately
    1.05
     Regardless
    1.04
     Sadly
    1.03
     Doing
    1.03
    Act Density 0.721%

    No Known Activations