© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 59623
    Prev
    Next
    INDEX
    Explanations

    numbers and counts

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

    The neuron is primarily detecting spelled-out number words (like “dos,” “tres,” “tolv,” “sep,” 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
    1
    -3.52
    0
    -3.09
    4
    -2.69
    '
    -2.67
    ↵↵
    -2.44
    [
    -2.44
    A
    -2.42
    -
    -2.39
    the
    -2.38
    8
    -2.36
    POSITIVE LOGITS
    ámica
    2.59
    ândia
    2.50
    );
    
    2.45
     शुर
    2.39
    二日
    2.38
     馆
    2.34
    ruzione
    2.31
     hundred
    2.31
    InstrumentedTest
    2.28
    棪
    2.27
    Activations Density 0.009%

    No Known Activations