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

    Neuronpedia

    INDEX
    Explanations

    conjunctions indicating contrast or opposition

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "However"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Balkans
    -0.69
    fell
    -0.61
     mid
    -0.60
     triangle
    -0.59
     award
    -0.57
     mids
    -0.57
     Flavoring
    -0.57
     Mountains
    -0.56
     cosmos
    -0.56
    mid
    -0.56
    POSITIVE LOGITS
    chers
    0.83
    ional
    0.81
    elsius
    0.78
    CLASSIFIED
    0.77
    lers
    0.75
    theless
    0.73
    ler
    0.73
    HER
    0.72
    antine
    0.70
     detractors
    0.69
    Act Density 0.034%

    No Known Activations