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

    Neuronpedia

    INDEX
    Explanations

    code, configurations, documentation

    np_max-act · gemini-2.0-flash

    This neuron detects comment header labels and metadata keywords (e.g. “MIME,” “Note,” “Reference”) in code comments.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     cottage
    -0.07
    modifiers
    -0.06
    ."""↵
    -0.06
    akedown
    -0.06
    (in
    -0.06
     rval
    -0.06
    _corr
    -0.06
    âng
    -0.06
    .');
↵
    -0.06
    _death
    -0.06
    POSITIVE LOGITS
     parentheses
    0.08
     resultat
    0.08
    alizace
    0.06
    فاق
    0.06
     Clause
    0.06
    strong
    0.06
    Coming
    0.06
     standings
    0.06
     renewed
    0.06
     Objective
    0.06
    Act Density 0.126%

    No Known Activations