© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 110500
    Prev
    Next
    INDEX
    Explanations

    package verification and cleanup

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

    section headings or titles in structured documents.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron activates strongly on numeric tokens (including integers, decimals, percentages, and version numbers).

    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
     "}";
    -1.60
     GEORGIA
    -1.50
     ",";
    -1.50
     PENNSYLVANIA
    -1.48
     '':
    -1.45
     ")";
    -1.44
    maxn
    -1.44
    '
    -1.43
    胯
    -1.42
     '';
    
    -1.41
    POSITIVE LOGITS
     of
    1.88
     as
    1.72
     brisk
    1.44
     \
    1.41
     opportunities
    1.39
     отношений
    1.39
     ответствен
    1.38
     geringen
    1.36
    year
    1.34
     größeren
    1.34
    Activations Density 0.097%

    No Known Activations