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

    Neuronpedia

    INDEX
    Explanations

    instances where the word "say" is used in various contexts or expressions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "say" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    xtap
    -0.78
    ockets
    -0.76
    mania
    -0.70
    pleting
    -0.65
    gotten
    -0.64
    olars
    -0.62
     Firearms
    -0.61
    ritional
    -0.61
     largeDownload
    -0.60
    mental
    -0.59
    POSITIVE LOGITS
     goodbye
    1.70
     hello
    1.33
     aloud
    1.21
     farewell
    1.14
     Goodbye
    1.07
     bye
    1.01
     hi
    1.01
     yes
    1.00
     loudly
    0.99
     sorry
    0.92
    Act Density 0.130%

    No Known Activations