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

    Neuronpedia

    INDEX
    Explanations

    mechanical parts

    np_max-act · gemini-2.0-flash

    This neuron detects technical terms naming mechanical parts and structures (e.g., locking, cam, tongue, stiffener, engagement, slanted section) in patent–style text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (right
    -0.07
    طار
    -0.07
     Bram
    -0.06
    itals
    -0.06
     Childhood
    -0.06
     Adoles
    -0.06
    Listing
    -0.06
    .lower
    -0.06
    razier
    -0.06
    ourney
    -0.06
    POSITIVE LOGITS
    .Sequential
    0.07
     Nicolas
    0.07
    ]+'
    0.07
     отнош
    0.06
    _TP
    0.06
    _Tick
    0.06
    لال
    0.06
     ngang
    0.06
     Json
    0.06
     protected
    0.06
    Act Density 0.069%

    No Known Activations