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

    Neuronpedia

    INDEX
    Explanations

    verbal mentions of specific information sources

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "cite."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    WINDOWS
    -0.70
     ALP
    -0.69
    normal
    -0.67
    OOOO
    -0.64
     SOLD
    -0.64
     Extras
    -0.63
     Ultra
    -0.63
     Shadows
    -0.61
    fighting
    -0.60
     Error
    -0.59
    POSITIVE LOGITS
     cite
    3.37
     cites
    2.51
     cited
    1.44
     citing
    1.40
     Citation
    1.30
     mention
    1.14
     recite
    1.08
     disproportionate
    1.04
     citations
    1.03
     mentions
    1.01
    Act Density 0.017%

    No Known Activations