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

    Neuronpedia

    INDEX
    Explanations

    adjectives describing depth

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "deep."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    icans
    -0.73
    orious
    -0.71
    ICAN
    -0.70
     AUTHOR
    -0.69
     WATCHED
    -0.69
    UID
    -0.69
    aurus
    -0.68
     Speedway
    -0.66
     MILL
    -0.65
    cules
    -0.65
    POSITIVE LOGITS
     vein
    1.01
    ened
    0.93
     deep
    0.91
     seeded
    0.88
     depth
    0.85
     penetration
    0.84
    deep
    0.82
    water
    0.79
     deepest
    0.78
     seated
    0.76
    Act Density 0.013%

    No Known Activations