© 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. 130575
    Prev
    Next
    INDEX
    Explanations

    The neuron is sensitive to numeric tokens (digits and numbers) in mathematical expressions.

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

    mathematical expressions with operators and numbers

    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
     that
    -1.52
     ガイド
    -1.24
     which
    -1.13
    În
    -1.12
     another
    -1.11
    terrasse
    -1.11
     skak
    -1.11
    無し
    -1.07
     カフェ
    -1.05
     afterwards
    -1.05
    POSITIVE LOGITS
    {};
    1.00
     deutschen
    0.99
     italienischen
    0.93
    表彰
    0.91
     منذ
    0.90
    托管
    0.89
    техни
    0.88
     Quando
    0.88
     ihrem
    0.88
    日向
    0.86
    Activations Density 0.205%

    No Known Activations