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

    Neuronpedia

    INDEX
    Explanations

    incomplete sentences/articles

    np_max-act · gemini-2.0-flash

    This neuron detects Italian verb inflections (especially third-person singular imperfect and subjunctive endings like ‑ava, ‑eva, ‑esse).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     symbolic
    -0.07
     doubled
    -0.07
     fans
    -0.07
    coll
    -0.07
     wis
    -0.06
    dont
    -0.06
     french
    -0.06
    safe
    -0.06
     toolkit
    -0.06
    pull
    -0.06
    POSITIVE LOGITS
     jsonArray
    0.07
    _BOTH
    0.06
     ATA
    0.06
     Dietary
    0.06
    0.06
    _admin
    0.06
    _dispatcher
    0.06
     नक
    0.06
    ظˆط
    0.06
     नगर
    0.06
    Act Density 0.278%

    No Known Activations