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

    Neuronpedia

    INDEX
    Explanations

    the word "extreme" as well as phrases containing the concept of extremity

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "extreme" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    shaw
    -0.83
    glas
    -0.80
    nance
    -0.80
    gin
    -0.79
    ËĪ
    -0.76
    olith
    -0.75
    bourg
    -0.73
    ************
    -0.71
    Reviewed
    -0.71
    bats
    -0.71
    POSITIVE LOGITS
     extremes
    1.14
     extreme
    1.11
    Extreme
    0.87
     Extreme
    0.86
     lengths
    0.82
     instances
    0.79
     temperatures
    0.78
    extreme
    0.77
     vetting
    0.76
     scenarios
    0.75
    Act Density 0.009%

    No Known Activations