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

    Neuronpedia

    INDEX
    Explanations

    -free

    np_max-act · gemini-2.0-flash

    This neuron activates on topical content words—concrete nouns and domain-specific terms (e.g., objects, places, activities, and technical or subject-specific vocabulary).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
    ,''
    -0.07
     erv
    -0.07
    ''
    -0.06
    ฤด
    -0.06
    .""
    -0.06
    ‎
    -0.06
    .''
    -0.06
    ��
    -0.06
    ufe
    -0.06
     ary
    -0.06
    POSITIVE LOGITS
    itemId
    0.08
    AccessType
    0.07
    �
    0.07
    ốt
    0.07
     حرکت
    0.07
    ONG
    0.07
     Scott
    0.07
    ожет
    0.07
    .TableName
    0.07
    Jean
    0.07
    Act Density 48.711%

    No Known Activations