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

    Neuronpedia

    INDEX
    Explanations

    words and phrases related to something being constricted or restricted, like "tight constraints" or "tight community."

    oai_token-act-pair · gpt-3.5-turbo

    repeated references to the concept of 'tightness'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ulhu
    -0.79
     Clever
    -0.77
     Twain
    -0.74
    oleon
    -0.69
     Courage
    -0.67
     Amph
    -0.66
    ista
    -0.65
     Resort
    -0.64
    ennial
    -0.64
     Lynd
    -0.64
    POSITIVE LOGITS
    ness
    1.06
    nesses
    1.01
     tight
    0.92
    heed
    0.83
     squeeze
    0.81
     fitting
    0.79
     throttle
    0.78
     toler
    0.76
    est
    0.76
    lining
    0.76
    Act Density 0.008%

    No Known Activations