© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 130832
    Prev
    Next
    INDEX
    Explanations

    Trudeau, Laos, Tianjin, Lesotho

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron detects capitalized named entities—proper nouns such as country and city names, personal names, and institutions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    '
    -3.19
     realizando
    -2.53
    </strong>
    -2.47
     That
    -2.39
     incorporar
    -2.34
    </h3>
    -2.30
    锇
    -2.28
    썽
    -2.27
    鍆
    -2.27
     että
    -2.23
    POSITIVE LOGITS
    轳
    2.81
    說
    2.42
    >
    2.38
    </b>
    2.33
    ad
    2.27
    數
    2.19
    蜑
    2.16
    ;</
    2.14
    注意事
    2.14
    鶲
    2.13
    Activations Density 0.004%

    No Known Activations