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

    Neuronpedia

    INDEX
    Explanations

    Quotation marks

    np_max-act · gemini-2.0-flash

    This neuron detects the opening quotation marks of string literals (the start of each `"…"` message) in the code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    不足
    -0.07
    methodVisitor
    -0.07
    лот
    -0.06
    $",
    -0.06
    でしょう
    -0.06
    ึก
    -0.06
    移动
    -0.06
    -equiv
    -0.06
     unlock
    -0.06
    ━━━━━━━━
    -0.06
    POSITIVE LOGITS
     hogy
    0.07
    apat
    0.07
     PBS
    0.07
     accession
    0.06
    atched
    0.06
    ningar
    0.06
    IGATION
    0.06
     arises
    0.06
    +'/
    0.06
     industries
    0.06
    Act Density 0.070%

    No Known Activations