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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    hardware/electronics-style alphanumeric identifiers and acronyms—such as model numbers, part codes, and versioned interface/spec tokens

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     embodiments
    -0.07
     depiction
    -0.07
     fifteen
    -0.07
    oltage
    -0.06
     Bew
    -0.06
     nineteen
    -0.06
     tape
    -0.06
     Dungeon
    -0.06
    species
    -0.06
    price
    -0.06
    POSITIVE LOGITS
    \core
    0.07
    اما
    0.07
    □□
    0.07
     Κ
    0.06
    ۳۶
    0.06
    enville
    0.06
     Schwartz
    0.06
     этот
    0.06
    �
    0.06
     Morav
    0.06
    Act Density 1.387%

    No Known Activations