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

    Neuronpedia

    INDEX
    Explanations

    personal pronouns followed by a verb in past tense

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific person

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    noon
    -0.70
    geries
    -0.67
     differential
    -0.65
    berra
    -0.63
    acters
    -0.62
     funer
    -0.61
     Powered
    -0.61
     locality
    -0.61
     quantity
    -0.60
     privile
    -0.60
    POSITIVE LOGITS
     joked
    1.19
     exclaimed
    1.18
     replied
    1.18
     said
    1.12
     remarked
    1.08
     wrote
    1.05
     tweeted
    1.02
     explained
    1.01
     says
    1.01
     laughed
    1.00
    Act Density 0.074%

    No Known Activations