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

    Neuronpedia

    INDEX
    Explanations

    references to a specific name mentioned multiple times throughout different contexts

    oai_token-act-pair · gpt-3.5-turbo

    words containing the sequence "ough."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    HOU
    -0.69
    vier
    -0.66
     Leviathan
    -0.63
     Babel
    -0.63
     Luthor
    -0.63
     Warlock
    -0.61
     Ibn
    -0.60
    âĹ¼
    -0.59
     Petra
    -0.59
     Hamm
    -0.58
    POSITIVE LOGITS
    ough
    0.91
    hesda
    0.88
    nesses
    0.87
    ness
    0.85
    sburgh
    0.81
    shots
    0.77
    tyard
    0.76
    é¾įåĸļ士
    0.75
    esy
    0.75
    ernels
    0.75
    Act Density 0.006%

    No Known Activations