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

    Neuronpedia

    INDEX
    Explanations

    research and technology

    np_max-act · gemini-2.0-flash

    This neuron fires on passive‐voice or definitional clauses—i.e. noun subjects occurring with a form of “be” (or modal “can be”) plus a past participle or descriptor.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ]).↵
    -0.07
    pred
    -0.06
    ']],
    -0.06
    /tmp
    -0.06
    ]];
    -0.06
     드립니다
    -0.06
     ：
    -0.06
    warnings
    -0.06
     เมตร
    -0.06
     Стар
    -0.06
    POSITIVE LOGITS
     cancelling
    0.07
     자동차
    0.07
     LOGGER
    0.07
    TOTYPE
    0.07
    ?=
    0.07
     pointing
    0.07
     CIS
    0.07
    执
    0.06
    asse
    0.06
    Resume
    0.06
    Act Density 0.414%

    No Known Activations