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

    Neuronpedia

    INDEX
    Explanations

    Importance/Success

    np_max-act · gemini-2.0-flash

    The neuron fires on abstract, nominalized concepts—gerunds or nouns that name processes, states, or qualities (e.g. “maintaining,” “balance,” “development”)—typically appearing as the head of a prepositional phrase.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    keywords related to web services and their discovery mechanisms, specifically in the context of grid computing.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ナー
    -0.06
    �
    -0.06
    encies
    -0.06
     nightmare
    -0.06
    ıf
    -0.06
     отверсти
    -0.06
    turnstile
    -0.06
     президент
    -0.06
     quadrant
    -0.06
     liegt
    -0.06
    POSITIVE LOGITS
     yelled
    0.07
    Reviews
    0.07
     Opcode
    0.07
    ILINE
    0.07
    AIR
    0.06
    =int
    0.06
    air
    0.06
     AIR
    0.06
    _encoder
    0.06
    ayi
    0.06
    Act Density 0.113%

    No Known Activations