© 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. 130742
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on numeric tokens (sequences of digits), in particular two‐digit numbers.

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

    Q: and why questions

    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
     this
    -1.20
    為に
    -1.10
     that
    -1.09
     enticing
    -0.96
    ;
    -0.95
    S
    -0.94
     which
    -0.93
    Note
    -0.92
     five
    -0.92
     Much
    -0.91
    POSITIVE LOGITS
    がありますが
    1.08
    では
    1.08
     écla
    1.05
    bten
    1.02
     prae
    1.01
     triste
    0.99
    extremely
    0.98
     naves
    0.98
    Nacional
    0.97
     ویکی‌پدیا
    0.96
    Activations Density 0.000%

    No Known Activations