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

    Neuronpedia

    INDEX
    Explanations

    words related to speech or speech-related activities

    oai_token-act-pair · gpt-3.5-turbo

    references to speech and communication

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Soldiers
    -0.70
     Royale
    -0.65
    astern
    -0.65
     Associ
    -0.63
    aro
    -0.63
    risome
    -0.62
    ño
    -0.62
    ellen
    -0.61
     Popular
    -0.61
     Jaguar
    -0.61
    POSITIVE LOGITS
    speech
    0.95
     speech
    0.94
     imped
    0.91
    writers
    0.86
    writer
    0.86
    writing
    0.80
    emic
    0.79
    anooga
    0.79
     uttered
    0.78
    ible
    0.78
    Act Density 0.023%

    No Known Activations