© 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. 130467
    Prev
    Next
    INDEX
    Explanations

    gravel or rubble

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

    The neuron strongly activates on mentions of granular solid materials and packing—particularly “gravel,” “packing,” and similar terms like “rubble.”

    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
    sandália
    -2.64
    '
    -2.58
    ↵↵
    -2.50
    ’
    -2.42
     ‘‘
    -2.33
     viaja
    -2.30
    瞜
    -2.28
    譟
    -2.27
     abhor
    -2.20
    焄
    -2.19
    POSITIVE LOGITS
    s
    3.31
    VolleyError
    2.31
    あります
    2.31
    心中的
    2.25
    婞
    2.16
    the
    2.11
    运气
    2.09
    ዡ
    2.09
    га
    2.09
    <bos>
    2.09
    Activations Density 0.002%

    No Known Activations