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

    Neuronpedia

    INDEX
    Explanations

    text related to outlining or presenting a structured plan or argument

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "laid out" or similar constructions indicating the presentation of ideas or plans

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    illus
    -0.71
    cius
    -0.68
     Bite
    -0.65
     Nickel
    -0.65
     unintentional
    -0.65
    BILITIES
    -0.64
    obos
    -0.62
     Mub
    -0.61
     acqu
    -0.61
    iets
    -0.60
    POSITIVE LOGITS
    posts
    0.93
     outlines
    0.87
    fitted
    0.81
    stretched
    0.78
    casts
    0.78
     groundwork
    0.76
    WER
    0.74
    cloth
    0.72
     outlining
    0.71
    ãģ¦
    0.69
    Act Density 0.032%

    No Known Activations