© 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. 104344
    Prev
    Next
    INDEX
    Explanations

    pockets for storage

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

    The neuron activates on mentions of storage features—words like “pocket,” “pockets,” “pouch,” “pouches,” or “compartments.”

    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
    The
    -2.23
     會
    -2.14
    梣
    -2.11
    on
    -2.08
    T
    -1.95
    鯓
    -1.94
    1
    -1.91
     邊
    -1.84
    .”
    -1.84
     料
    -1.82
    POSITIVE LOGITS
     '
    2.52
     unſ
    2.27
    colgante
    2.05
    ୩
    1.99
    mäs
    1.98
     serão
    1.97
    跸
    1.95
     pockets
    1.91
    蕻
    1.90
    因为他
    1.89
    Activations Density 0.002%

    No Known Activations