© 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. 11058
    Prev
    Next
    INDEX
    Explanations

    code snippets and identifiers

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

    The neuron flags numeric tokens—literal numbers and numerals—in code or text.

    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
     of
    -1.17
     Bestimmungen
    -1.02
    蜑
    -0.99
    מְ
    -0.99
    מבר
    -0.96
    theim
    -0.96
     discours
    -0.94
    Vaata
    -0.94
    ibilidade
    -0.93
    身高
    -0.92
    POSITIVE LOGITS
     fortfarande
    1.09
     =>
    1.02
    --
    1.01
     опять
    0.99
     senare
    0.98
    ğimiz
    0.98
     ändå
    0.96
     різних
    0.93
     ещё
    0.92
     kanske
    0.92
    Activations Density 0.007%

    No Known Activations