© 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. 60965
    Prev
    Next
    INDEX
    Explanations

    code snippets, programming language constructs

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

    The neuron fires on numeric literal tokens (both integers and decimals) in the text.

    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
     perioden
    -1.86
     flera
    -1.81
     dagens
    -1.73
     denna
    -1.70
     februari
    -1.64
     ۳
    -1.63
    ۹
    -1.60
     tanken
    -1.57
     problemet
    -1.57
     =$
    -1.57
    POSITIVE LOGITS
     they
    2.38
     when
    2.34
     on
    2.33
     at
    2.33
     it
    2.23
     or
    2.19
     which
    1.99
     there
    1.99
     chose
    1.90
     from
    1.81
    Activations Density 0.044%

    No Known Activations