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

    Neuronpedia

    INDEX
    Explanations

    references to deceased individuals

    oai_token-act-pair · gpt-3.5-turbo

    references to deceased individuals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ebook
    -0.83
    cking
    -0.74
    ifling
    -0.71
    acle
    -0.71
    athing
    -0.70
    aly
    -0.70
    plex
    -0.70
    agra
    -0.69
    elling
    -0.68
    ramid
    -0.68
    POSITIVE LOGITS
     deceased
    1.04
     persons
    0.81
     comrades
    0.73
     sins
    0.73
     occupants
    0.72
     fug
    0.72
     fugitive
    0.71
     seniors
    0.70
    etheless
    0.70
     relatives
    0.70
    Act Density 0.027%

    No Known Activations