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

    Neuronpedia

    INDEX
    Explanations

    proper nouns followed by specific details or entities

    oai_token-act-pair · gpt-3.5-turbo

    the word "the" and other references indicating a focus on entities and subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Parables
    -0.67
     discont
    -0.66
     Petition
    -0.60
     Wire
    -0.59
     Message
    -0.59
    10000
    -0.55
     certs
    -0.55
     constantly
    -0.55
     Depth
    -0.55
     Rated
    -0.55
    POSITIVE LOGITS
     sexes
    1.01
     nor
    0.96
     genders
    0.75
    "]=>
    0.72
    alde
    0.67
     statically
    0.66
    ãĥ¬
    0.65
    %]
    0.65
    isSpecialOrderable
    0.64
    fml
    0.64
    Act Density 0.170%

    No Known Activations