© 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. 128241
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numeric tokens and numbered labels (e.g., steps, days, chapter or group numbers).

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

    numbers following keywords or as dates

    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
     from
    -1.28
     new
    -1.23
     Canadá
    -1.20
     found
    -1.14
     that
    -1.13
     of
    -1.12
    »).
    -1.12
    devServer
    -1.09
    elsif
    -1.08
     consulté
    -1.07
    POSITIVE LOGITS
    3
    1.53
    8
    1.50
    ha
    1.48
    6
    1.44
    5
    1.43
    7
    1.41
    ar
    1.34
    1
    1.34
     alcune
    1.33
    wa
    1.30
    Activations Density 0.001%

    No Known Activations