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

    Neuronpedia

    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash

    This neuron detects the patent-style introductory phrase where the text “relates to” (often preceded by “more particularly” or “the present invention”) specifies the invention’s subject.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     representa
    -0.07
     autumn
    -0.07
    /todo
    -0.07
    ----------------------------
    -0.07
     WORK
    -0.07
     undercut
    -0.07
    forget
    -0.07
     Frozen
    -0.06
     Tips
    -0.06
    ursor
    -0.06
    POSITIVE LOGITS
    <State
    0.06
     Ї
    0.06
    _sq
    0.06
    ([$
    0.06
    .money
    0.06
    ayla
    0.05
     магаз
    0.05
     puppies
    0.05
    ленный
    0.05
     dane
    0.05
    Act Density 0.015%

    No Known Activations