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

    Neuronpedia

    INDEX
    Explanations

    verbs that are used to describe or explain something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "describe" and its variations, indicating a focus on explanations or descriptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    amus
    -0.61
     Hav
    -0.61
     encro
    -0.60
     innocent
    -0.60
     ster
    -0.59
     forfe
    -0.58
    itor
    -0.58
    eful
    -0.58
     revenue
    -0.58
     stead
    -0.57
    POSITIVE LOGITS
     describe
    3.54
     describes
    2.18
     describing
    2.03
     characterize
    1.98
     describ
    1.89
     summarize
    1.77
     explain
    1.70
     depict
    1.60
     described
    1.59
     define
    1.57
    Act Density 0.009%

    No Known Activations