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

    Neuronpedia

    INDEX
    Explanations

    references to celebrities

    oai_token-act-pair · gpt-3.5-turbo

    references to celebrities and celebrity culture

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ¼
    -0.86
    anus
    -0.86
    choes
    -0.85
    hematic
    -0.80
    ²¾
    -0.78
    ¾
    -0.78
    THER
    -0.75
    ¸
    -0.74
    Ķ
    -0.73
    tered
    -0.72
    POSITIVE LOGITS
    rities
    1.12
     endors
    0.98
     chef
    0.98
     endorsements
    0.96
     gossip
    0.95
     chefs
    0.88
     celebrities
    0.83
     nude
    0.82
     celebrity
    0.77
     athlete
    0.77
    Act Density 0.036%

    No Known Activations