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

    Neuronpedia

    INDEX
    Explanations

    sentences starting with "It is" followed by an adjective and a verb

    oai_token-act-pair · gpt-3.5-turbo

    repetitions of the phrase "it is" to emphasize statements or assertions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    bies
    -0.79
     Bung
    -0.60
     Telescope
    -0.60
     Previous
    -0.58
    iling
    -0.58
     Lack
    -0.58
     coasts
    -0.58
     accuses
    -0.56
     Want
    -0.56
     wishes
    -0.56
    POSITIVE LOGITS
    nt
    1.02
     unlikely
    1.01
     definitely
    0.98
    rael
    0.96
     not
    0.89
     abundantly
    0.87
     hereby
    0.87
     unclear
    0.87
     easier
    0.86
     also
    0.85
    Act Density 0.418%

    No Known Activations