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

    Neuronpedia

    INDEX
    Explanations

    references to the physical space of a classroom

    oai_token-act-pair · gpt-3.5-turbo

    repeated references to classrooms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    alez
    -0.81
    atra
    -0.77
    ulf
    -0.75
    ods
    -0.75
    atom
    -0.72
    berto
    -0.70
    ome
    -0.70
    onement
    -0.69
    olid
    -0.68
    omes
    -0.67
    POSITIVE LOGITS
     classroom
    0.84
     classrooms
    0.83
     Teachers
    0.82
    room
    0.78
     Teaching
    0.77
     Teacher
    0.74
     Corner
    0.73
    LAB
    0.72
    girls
    0.72
    inary
    0.71
    Act Density 0.011%

    No Known Activations