© 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. 131019
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on stand-alone numeric literals (digit sequences) in code or text.

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

    logging and printing

    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
     to
    -1.29
    риса
    -1.13
    Jahre
    -1.10
     апель
    -1.10
    Podob
    -1.08
    ₩
    -1.08
     biela
    -1.08
    ritsar
    -1.07
    lineTo
    -1.06
     criaturas
    -1.06
    POSITIVE LOGITS
    ----------------
    1.20
    b
    1.07
    th
    1.05
    是以
    1.00
    囉
    0.99
    h
    0.99
    cillary
    0.97
     frecuentemente
    0.94
    li
    0.94
     Ali
    0.94
    Activations Density 0.001%

    No Known Activations