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

    Neuronpedia

    INDEX
    Explanations

    programming forums

    np_max-act · gemini-2.0-flash

    This neuron activates on plural resource nouns in API-style questions or paths (e.g. “requests,” “posts,” “issues,” “tweets,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (Test
    -0.07
     indice
    -0.07
    (ph
    -0.07
     mgr
    -0.07
     analyse
    -0.06
    BODY
    -0.06
     Fiction
    -0.06
     blue
    -0.06
     evidence
    -0.06
     ramifications
    -0.06
    POSITIVE LOGITS
    娘
    0.07
    _background
    0.06
     mẽ
    0.06
    ким
    0.06
    ulia
    0.06
    ingo
    0.06
     affiliate
    0.06
    _kelas
    0.06
    .clearRect
    0.06
     الإن
    0.06
    Act Density 0.037%

    No Known Activations