© 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. 22-GEMMASCOPE-RES-131K
    7. 130955
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively activates on standalone numeric tokens or numeric literals (sequences of digits) in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    version numbers and package declarations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     our
    -1.80
     what
    -1.25
     when
    -1.16
     several
    -1.15
     new
    -1.15
     this
    -1.13
     any
    -1.13
     two
    -1.12
     some
    -1.11
     larger
    -1.10
    POSITIVE LOGITS
    Harga
    1.30
    ഇ
    1.20
    windig
    1.18
    இ
    1.18
     normalt
    1.14
    
    1.13
    WriteLiteral
    1.11
    شما
    1.10
    Vendredi
    1.10
    Tahun
    1.09
    Activations Density 0.004%

    No Known Activations