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

    Neuronpedia

    INDEX
    Explanations

    instances of emphasis, contrast, or unexpectedness

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "even."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rend
    -0.79
    asus
    -0.78
    aim
    -0.76
    cel
    -0.76
    orts
    -0.71
    ffen
    -0.69
    uble
    -0.69
    lli
    -0.68
    ENTION
    -0.67
    cent
    -0.66
    POSITIVE LOGITS
     though
    1.49
     tho
    1.13
     if
    1.04
    handedly
    0.95
    though
    0.84
    handed
    0.83
     remotely
    0.80
     amidst
    0.78
     assuming
    0.75
     accounting
    0.75
    Act Density 0.040%

    No Known Activations