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

    Neuronpedia

    INDEX
    Explanations

    Multiple choice questions

    np_max-act · gemini-2.0-flash

    This neuron activates on multiple-choice option markers—specifically the parentheses and letters (A, B, C, D) used to label answer choices.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    英语
    -0.06
    _task
    -0.06
    izzie
    -0.06
     Manor
    -0.06
     Jar
    -0.06
     آینده
    -0.06
    Aaron
    -0.06
    .manual
    -0.06
     لكرة
    -0.06
    Common
    -0.06
    POSITIVE LOGITS
    letal
    0.08
     Marks
    0.07
     -:
    0.07
    _initial
    0.07
    。',↵
    0.06
    (global
    0.06
    IZATION
    0.06
    -security
    0.06
    tection
    0.06
    dq
    0.06
    Act Density 0.006%

    No Known Activations