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

    Neuronpedia

    INDEX
    Explanations

    phrases related to extended periods of performance or activity

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of a "run" or ongoing sequence of events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Voice
    -0.74
     Males
    -0.71
     Hots
    -0.70
     Photographer
    -0.67
    omical
    -0.66
    ĨĴ
    -0.65
     Virtue
    -0.65
    oshop
    -0.64
    hammad
    -0.62
     Canad
    -0.62
    POSITIVE LOGITS
    aways
    1.07
    escape
    1.04
    gs
    1.02
    runner
    0.99
    nings
    0.98
    swick
    0.98
    ways
    0.94
    ners
    0.93
    gression
    0.92
    times
    0.91
    Act Density 0.033%

    No Known Activations