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

    Neuronpedia

    INDEX
    Explanations

    adjectives that express varying degrees of positivity or negativity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     chron
    -0.15
    daily
    -0.15
     Ú¯ÛĮر
    -0.14
    aken
    -0.14
    δη
    -0.14
    ohan
    -0.13
    daÅŁ
    -0.13
    637
    -0.13
    utting
    -0.13
    486
    -0.13
    POSITIVE LOGITS
     few
    0.35
     week
    0.33
     month
    0.29
    few
    0.28
     year
    0.28
     couple
    0.28
     start
    0.27
     Few
    0.26
    Few
    0.26
     time
    0.25
    Act Density 0.095%

    No Known Activations