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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 8867
    Prev
    Next
    INDEX
    Explanations

    names introducing factual statements

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

    The neuron responds to very short, common function words (two‐ or three‐letter tokens such as “an,” “du,” “man,” “nu,” “he,” etc.) across different languages.

    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
    鴣
    -2.80
    的新
    -2.69
    沺
    -2.31
     sudadera
    -2.28
    的高
    -2.25
     جدید
    -2.19
    🪸
    -2.19
    鼇
    -2.17
     UNIVE
    -2.17
    的一个
    -2.17
    POSITIVE LOGITS
    \
    2.31
    </td>
    2.25
     ثلاثة
    2.22
    despite
    2.19
     It
    2.09
     ‘‘
    2.05
    行业的
    2.05
     ‘
    2.05
     to
    2.02
    それに
    1.98
    Activations Density 0.000%

    No Known Activations