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

    Neuronpedia

    INDEX
    Explanations

    phrases with colons used for emphasis or listing

    oai_token-act-pair · gpt-3.5-turbo

    punctuation and quotation marks used in speech or dialogue

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Developer
    -0.80
    duc
    -0.62
    Room
    -0.61
     giveaway
    -0.59
    Cla
    -0.59
    Beg
    -0.58
    WOOD
    -0.57
    starter
    -0.56
    Article
    -0.56
    arters
    -0.55
    POSITIVE LOGITS
    aurus
    0.76
    nces
    0.69
    ãĤª
    0.69
    xual
    0.68
    ension
    0.68
    ensions
    0.67
    inctions
    0.66
    iband
    0.66
    ional
    0.65
     Flavoring
    0.65
    Act Density 0.105%

    No Known Activations