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

    Neuronpedia

    INDEX
    Explanations

    niece

    np_max-act · gemini-2.0-flash

    The neuron fires on rarer, multi-syllable specialized terms (often compound or hyphenated) that appear in technical, scientific, or niche contexts.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Benef
    -0.07
    (up
    -0.07
    inh
    -0.06
    	ep
    -0.06
    (the
    -0.06
     &&↵
    -0.06
    、私
    -0.06
     lot
    -0.06
     ↵  ↵
    -0.06
     verifies
    -0.06
    POSITIVE LOGITS
     Україн
    0.07
    ENTION
    0.07
     emploi
    0.07
     广
    0.07
    -tm
    0.06
     požad
    0.06
     Partner
    0.06
     lĩnh
    0.06
     nghiêm
    0.06
     utiliser
    0.06
    Act Density 0.085%

    No Known Activations