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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a comparison or contrast between two ideas

    oai_token-act-pair · gpt-3.5-turbo

    transitional phrases indicating contrast or comparison

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    IQ
    -0.70
     UNCLASSIFIED
    -0.63
    stead
    -0.60
    DL
    -0.58
    ilty
    -0.58
     Estimated
    -0.55
     compr
    -0.55
    GPU
    -0.54
    co
    -0.54
    coat
    -0.53
    POSITIVE LOGITS
     relying
    0.74
     let
    0.72
    SPONSORED
    0.70
     ours
    0.66
    mere
    0.64
     Instead
    0.61
     opting
    0.59
    invoke
    0.59
     recomb
    0.58
     capit
    0.58
    Act Density 0.088%

    No Known Activations