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

    Neuronpedia

    INDEX
    Explanations

    Essays

    np_max-act · gemini-2.0-flash

    This neuron detects French-language tokens associated with academic literary analysis (e.g., parts of words like “litt ”, “éra ”, “thèmes,” “structure,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    oklyn
    -0.07
    LIN
    -0.06
    BMI
    -0.06
     Valent
    -0.06
     Stoke
    -0.06
    _RULE
    -0.06
    &)
    -0.06
    .sec
    -0.06
    paged
    -0.06
    Na
    -0.06
    POSITIVE LOGITS
    pub
    0.06
     قال
    0.06
    ，可以
    0.06
    �
    0.06
    wahl
    0.06
     الحرب
    0.06
    ρχ
    0.06
     deb
    0.06
     incorporating
    0.06
    (sentence
    0.06
    Act Density 0.064%

    No Known Activations