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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    This neuron detects German technical procedure terms used as button labels or identifiers—e.g. “Öl prüfen,” “Inspektion,” and “Abnahme.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     yans
    -0.07
    ・・
    -0.07
    inks
    -0.06
    ặng
    -0.06
    符
    -0.06
    .GetEnumerator
    -0.06
     logits
    -0.06
    ..'
    -0.06
    .Register
    -0.06
    setattr
    -0.06
    POSITIVE LOGITS
    ]]></
    0.07
    cerpt
    0.07
    ,date
    0.06
     worrying
    0.06
     suspect
    0.06
     ",";↵
    0.06
    念
    0.06
    ите
    0.06
     thai
    0.06
    _EVENT
    0.06
    Act Density 0.258%

    No Known Activations