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

    Neuronpedia

    INDEX
    Explanations

    references to roommates and dormitories

    oai_token-act-pair · gpt-3.5-turbo

    references to roommates and dorm-related experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Spear
    -0.79
     arte
    -0.78
    Chain
    -0.74
    ifact
    -0.72
    Force
    -0.71
    ingers
    -0.69
     stones
    -0.69
    asive
    -0.69
    force
    -0.68
     Insight
    -0.67
    POSITIVE LOGITS
     roomm
    3.10
     roommate
    3.01
     dorm
    2.57
     classmate
    1.49
     classmates
    1.29
     cowork
    1.28
     coworkers
    1.25
     motel
    1.20
     squirrel
    1.16
     bunk
    1.10
    Act Density 0.041%

    No Known Activations