© 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. 101914
    Prev
    Next
    INDEX
    Explanations

    punctuation and conjunctions

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

    The neuron fires strongly on numeric tokens (including decimals, percentages, and other figures).

    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
     who
    -1.91
     comprise
    -1.51
     siempre
    -1.48
    шении
    -1.44
    崛起
    -1.40
    lässlich
    -1.39
    孳
    -1.38
    delsen
    -1.37
     ALWAYS
    -1.34
     ј
    -1.31
    POSITIVE LOGITS
    ↵
    1.70
     or
    1.48
     in
    1.44
     two
    1.33
     for
    1.31
    or
    1.29
    $
    1.27
     over
    1.23
     Today
    1.21
     out
    1.20
    Activations Density 0.229%

    No Known Activations