© 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. 65186
    Prev
    Next
    INDEX
    Explanations

    Wan and WAN network connectivity

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

    This neuron fires on tokens containing the letter sequence “wan,” whether as part of proper names (e.g. “Wan”) or the acronym “WAN.”

    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
     gustado
    -2.88
    ⧼
    -2.55
    Nazwa
    -2.53
    怼
    -2.50
    Примеча
    -2.48
    涐
    -2.47
     policia
    -2.42
    开的
    -2.42
    镚
    -2.42
    ’,
    -2.38
    POSITIVE LOGITS
    $
    3.31
    [
    3.20
    )
    3.09
    (
    2.98
    9
    2.94
    You
    2.92
    i
    2.89
    8
    2.81
    ised
    2.73
    or
    2.69
    Activations Density 0.005%

    No Known Activations