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

    Neuronpedia

    INDEX
    Explanations

    phrases related to statements or beliefs being passed on or circulated, often involving trust or credibility

    oai_token-act-pair · gpt-3.5-turbo

    phrases that emphasize the concept of "word," particularly in the context of testimony and communication

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Dise
    -0.73
    MRI
    -0.72
    ipment
    -0.67
     volt
    -0.67
    iaries
    -0.64
     Steps
    -0.64
    iencies
    -0.64
    ngth
    -0.64
     ponds
    -0.62
     Thermal
    -0.61
    POSITIVE LOGITS
     aloud
    0.88
     phrases
    0.86
     Arabic
    0.82
     uttered
    0.79
     poem
    0.76
     pronounced
    0.76
     acronym
    0.76
     spoken
    0.76
     abbre
    0.74
     curses
    0.73
    Act Density 0.282%

    No Known Activations