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

    Neuronpedia

    INDEX
    Explanations

    words related to speaking or the ability to speak

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "speak" and its variations, often in the context of communication difficulties or language barriers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    aredevil
    -0.76
    collection
    -0.72
    \/\/
    -0.69
    igor
    -0.67
     supplying
    -0.66
    intendent
    -0.66
    rw
    -0.65
    folio
    -0.65
     Gamble
    -0.64
    rolet
    -0.64
    POSITIVE LOGITS
     aloud
    1.18
     louder
    1.09
     english
    1.06
     fluent
    1.03
     English
    0.95
     loudly
    0.95
     spoken
    0.93
     voices
    0.92
     Arabic
    0.87
     phon
    0.87
    Act Density 0.091%

    No Known Activations