© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 15020
    Prev
    Next
    INDEX
    Explanations

    code formatting punctuation

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron activates on numeric tokens or digit sequences (e.g., version numbers, integers, or regex patterns capturing digits).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     ytter
    -1.18
    Новые
    -1.14
     مشکلات
    -1.10
    Semua
    -1.09
    Книга
    -1.09
     новые
    -1.09
    Jangan
    -1.04
     pesta
    -1.02
    pography
    -1.02
     only
    -1.02
    POSITIVE LOGITS
      
    1.06
     outskirts
    0.99
     k
    0.98
     teilen
    0.97
    ."
    0.96
     Show
    0.96
     Lower
    0.96
     okay
    0.95
    th
    0.93
     jedenfalls
    0.93
    Activations Density 0.014%

    No Known Activations