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

    Neuronpedia

    INDEX
    Explanations

    physical body parts like eyes, hands, lips, cheeks, and specific actions involving them

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    arenthood
    -0.82
     Rohing
    -0.81
    auri
    -0.81
     Byrne
    -0.73
     Yugoslavia
    -0.70
     Worldwide
    -0.70
     Haku
    -0.69
    humane
    -0.68
     Zah
    -0.67
     Heller
    -0.66
    POSITIVE LOGITS
     dipped
    0.93
     smelled
    0.89
     stained
    0.88
     burned
    0.88
     slipped
    0.87
     froze
    0.87
     dyed
    0.86
     crossed
    0.86
     cane
    0.84
     crawled
    0.83
    Act Density 8.899%

    No Known Activations