© 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. 130825
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on multi‐digit numeric tokens—especially years, dates, and other multi-digit numbers.

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

    years and dates

    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
    @@@@@@@@
    -1.24
    magine
    -1.20
    
    -1.17
     déroule
    -1.16
     geste
    -1.15
    @@@
    -1.14
     唯美
    -1.14
     GRATIS
    -1.13
     tranſ
    -1.13
    despite
    -1.13
    POSITIVE LOGITS
     countless
    1.41
     some
    1.36
     various
    1.30
     D
    1.29
     Some
    1.24
     those
    1.22
     even
    1.21
     innumerable
    1.21
     Very
    1.20
     any
    1.18
    Activations Density 0.016%

    No Known Activations