© 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. 130878
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on numeric tokens—especially measurements, statistics, dollar figures, and percentages.

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

    just over or under

    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
    㩒
    -1.34
    小孩子
    -1.27
     pracuje
    -1.27
     Gorbachev
    -1.23
     وكل
    -1.21
    Skocz
    -1.21
    íguez
    -1.20
    ጵ
    -1.20
     desbloquear
    -1.19
    职场
    -1.19
    POSITIVE LOGITS
    ʽ
    1.13
    u
    1.05
     Of
    1.04
    微微
    1.02
    Slightly
    0.99
     entiendo
    0.98
    v
    0.97
    d
    0.95
    b
    0.94
    w
    0.94
    Activations Density 0.015%

    No Known Activations