© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 68202
    Prev
    Next
    INDEX
    Explanations

    foreign languages and abbreviations

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

    This neuron fires on tokens from non-English text—especially those containing diacritics or other non-ASCII letters.

    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
    After
    -1.63
    Only
    -1.59
    ٨
    -1.55
    They
    -1.50
    鎳
    -1.44
    三个
    -1.43
    Once
    -1.42
     épa
    -1.42
    Although
    -1.40
    Despite
    -1.40
    POSITIVE LOGITS
     such
    1.88
     of
    1.82
     also
    1.82
     then
    1.64
     against
    1.62
     including
    1.57
     strategy
    1.50
     businesses
    1.49
     these
    1.49
     according
    1.48
    Activations Density 0.037%

    No Known Activations