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

    Neuronpedia

    INDEX
    Explanations

    probabilistic language indicating potential outcomes or scenarios

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ê±´
    -0.16
     geil
    -0.15
     Couldn
    -0.15
    reta
    -0.15
    Å¡lo
    -0.14
    Couldn
    -0.14
    uppen
    -0.14
    ç»Īäºİ
    -0.14
    macen
    -0.14
     finally
    -0.14
    POSITIVE LOGITS
     sometimes
    0.38
     often
    0.34
    sometimes
    0.31
     oft
    0.29
     Sometimes
    0.28
    often
    0.28
     seem
    0.28
    Sometimes
    0.27
     range
    0.26
    ometimes
    0.24
    Act Density 0.074%

    No Known Activations