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

    Neuronpedia

    INDEX
    Explanations

    pronoun

    np_max-act · gemini-2.0-flash

    text related to instructions, setup, guidance, or descriptions of AI behaviors.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron selectively activates on the pronoun “it” (in both lowercase and uppercase forms).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    veloper
    -0.08
     Friends
    -0.06
     doctors
    -0.06
     budd
    -0.06
     leaders
    -0.06
     й
    -0.06
     comed
    -0.06
     bullying
    -0.06
     volunteering
    -0.06
    orrent
    -0.06
    POSITIVE LOGITS
     batchSize
    0.08
     searching
    0.07
     матері
    0.06
    柴
    0.06
    {}]
    0.06
     toward
    0.06
    	vertex
    0.06
    雪
    0.06
    (.)
    0.06
    없이
    0.06
    Act Density 0.005%

    No Known Activations