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

    Neuronpedia

    INDEX
    Explanations

    instances where the phrase "That is to say" is used

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ersed
    -0.76
    oval
    -0.75
    res
    -0.69
    breaks
    -0.64
    ĪĴ
    -0.63
    urg
    -0.63
    rir
    -0.61
    rollers
    -0.60
     awaited
    -0.60
    swick
    -0.60
    POSITIVE LOGITS
     say
    1.29
     paraph
    0.91
     SAY
    0.86
     quote
    0.85
     diminish
    0.84
     wit
    0.83
     mention
    0.80
     deny
    0.79
    Say
    0.78
     imply
    0.75
    Act Density 0.035%

    No Known Activations