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

    Neuronpedia

    INDEX
    Explanations

    conditional statements with a cause-effect relationship

    oai_token-act-pair · gpt-3.5-turbo

    conditional statements that imply a contrast or challenge to a preceding assertion

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    pec
    -0.72
    olitical
    -0.70
    VERTISEMENT
    -0.68
    ouble
    -0.67
    oked
    -0.66
    nai
    -0.66
    apsed
    -0.66
    resent
    -0.64
    ThumbnailImage
    -0.64
    minster
    -0.64
    POSITIVE LOGITS
     then
    1.31
    then
    1.29
     surely
    1.12
     why
    1.10
     chances
    1.04
     THEN
    1.04
     congratulations
    0.96
     please
    0.91
     maybe
    0.88
     it
    0.87
    Act Density 0.220%

    No Known Activations