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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron is responsive to numeric literal tokens—especially immediate values, addresses, and constants in assembly‐style code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Not
    -0.07
    qd
    -0.07
     Not
    -0.06
     swear
    -0.06
     pago
    -0.06
    ۱۴
    -0.06
    "No
    -0.06
    .Abs
    -0.06
    sh
    -0.06
    ButtonText
    -0.06
    POSITIVE LOGITS
     Branch
    0.07
    .check
    0.06
     accumulated
    0.06
     dazz
    0.06
     objet
    0.06
     experimented
    0.06
    oord
    0.06
    lac
    0.06
    جة
    0.06
     objekt
    0.06
    Act Density 0.011%

    No Known Activations