© 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. 10-GEMMASCOPE-RES-131K
    7. 100382
    Prev
    Next
    INDEX
    Explanations

    numbers and identifiers

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

    The neuron strongly responds to numeric tokens (especially the digit “1” as it appears in years, numbered lists, and similar contexts).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     manquer
    -2.50
    锱
    -2.47
    躂
    -2.33
    
    -2.30
     piaci
    -2.30
     geschenk
    -2.28
    
    -2.22
     dezelve
    -2.20
     protéines
    -2.19
     chaînes
    -2.17
    POSITIVE LOGITS
     of
    2.50
     -
    2.33
     Bedürf
    2.25
     Umfeld
    2.23
     Ankunft
    2.13
     Auffassung
    2.11
     Aussehen
    2.11
    2
    2.09
     since
    2.09
     '
    2.08
    Activations Density 0.034%

    No Known Activations