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

    Neuronpedia

    INDEX
    Explanations

    the phrase "in a row"

    oai_token-act-pair · gpt-3.5-turbo

    references to consecutive or repeated sequences termed as "rows."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    allo
    -0.73
    acea
    -0.72
    Alpha
    -0.67
    antle
    -0.63
     Cutler
    -0.62
    zin
    -0.62
    essential
    -0.62
    ovsky
    -0.61
    bringer
    -0.60
    obin
    -0.60
    POSITIVE LOGITS
     row
    4.02
     rows
    2.63
     Row
    2.22
    Row
    2.15
    row
    1.64
     column
    1.36
     Column
    1.20
    Column
    1.16
     columns
    1.12
    rows
    1.08
    Act Density 0.019%

    No Known Activations