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

    Neuronpedia

    INDEX
    Explanations

    his

    np_max-act · gemini-2.0-flash

    This neuron fires on the “backbone” of simple clauses—marking the subject (pronoun or noun), its auxiliary/main verb (and often the object pronoun)—i.e. the core S-V(-O) structure of a sentence.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    undef
    -0.06
    GLenum
    -0.06
    Pragma
    -0.06
     decom
    -0.06
    CCR
    -0.06
     GL
    -0.06
     inventor
    -0.06
     valor
    -0.06
    िथ
    -0.06
    받
    -0.06
    POSITIVE LOGITS
     lengthy
    0.07
    说的
    0.07
     }?>↵
    0.07
    seq
    0.07
    '/>↵
    0.07
     proteins
    0.07
     planting
    0.06
    pr
    0.06
     radios
    0.06
    %↵↵
    0.06
    Act Density 0.030%

    No Known Activations