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

    Neuronpedia

    INDEX
    Explanations

    phrases related to legal matters and humor

    oai_token-act-pair · gpt-3.5-turbo

    special characters or symbols used in text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     scatter
    -0.73
     dirt
    -0.61
     scattering
    -0.61
     detached
    -0.60
     cyan
    -0.60
     maxim
    -0.58
     minim
    -0.57
     staggered
    -0.56
    wagen
    -0.55
     weighted
    -0.55
    POSITIVE LOGITS
    ¹
    0.93
    £
    0.93
    º
    0.86
    ¢
    0.85
    âĢł
    0.84
    âĹ¼
    0.84
    į
    0.83
    ¡
    0.83
    ISIS
    0.80
    ¼
    0.79
    Act Density 0.607%

    No Known Activations