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

    Neuronpedia

    INDEX
    Explanations

    instances where quantities are specified in comparison to a larger number

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "more."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ĺħ
    -0.92
    xtap
    -0.75
     Recording
    -0.72
     Shooter
    -0.71
    imaru
    -0.67
     Record
    -0.66
     Difference
    -0.65
    IDS
    -0.62
    UCK
    -0.62
     Passage
    -0.61
    POSITIVE LOGITS
     than
    1.16
     elaborate
    1.03
     complicated
    1.02
     sophisticated
    0.99
     stringent
    0.96
     manageable
    0.96
     expensive
    0.95
     extensive
    0.95
     detailed
    0.93
     nuanced
    0.91
    Act Density 0.116%

    No Known Activations