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

    Neuronpedia

    INDEX
    Explanations

    questions or prompts indicating a searching or probing nature

    oai_token-act-pair · gpt-3.5-turbo

    rhetorical questions or inquiries that provoke thought

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    roud
    -0.64
    ographed
    -0.63
     earthqu
    -0.61
    itton
    -0.59
    ICT
    -0.58
    shaw
    -0.57
    æĪ¦
    -0.56
    aditional
    -0.56
    ãĤ§
    -0.54
    RAG
    -0.54
    POSITIVE LOGITS
     Does
    1.27
     Why
    1.24
     Which
    1.15
     why
    1.14
     Should
    1.12
     What
    1.11
     Who
    1.10
     Are
    1.10
     Would
    1.10
     How
    1.09
    Act Density 0.121%

    No Known Activations