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

    Neuronpedia

    INDEX
    Explanations

    proper names of people, places, and organizations

    oai_token-act-pair · gpt-3.5-turbo

    proper nouns and specific names related to individuals, organizations, or notable entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Saying
    -0.93
     saying
    -0.78
    soDeliveryDate
    -0.74
     mentioning
    -0.72
    actionDate
    -0.68
    soType
    -0.67
     Talking
    -0.66
     stating
    -0.66
     confirming
    -0.65
     messing
    -0.64
    POSITIVE LOGITS
     boasts
    1.17
     embodies
    1.13
     strives
    1.09
     possesses
    1.07
     nevertheless
    1.04
     delivers
    1.04
     relies
    1.01
     seeks
    1.00
     enjoys
    0.99
     resembles
    0.98
    Act Density 0.327%

    No Known Activations