© 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. 131060
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on individual numeric digit tokens (e.g. parts of phone numbers).

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

    common word pairs

    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
    bote
    -1.08
    ∅
    -1.05
    noo
    -1.04
     निम्नलिखित
    -1.00
    FEB
    -0.99
    ⌈
    -0.98
    nora
    -0.98
     февраля
    -0.96
    CAA
    -0.96
    PLUS
    -0.95
    POSITIVE LOGITS
     and
    1.19
     one
    1.16
    Belum
    1.13
    Примеча
    1.06
    Makanan
    1.05
    Какво
    1.04
    drei
    1.01
    ？（
    1.01
    Pener
    0.98
    توضیحات
    0.97
    Activations Density 0.000%

    No Known Activations