© 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. 38939
    Prev
    Next
    INDEX
    Explanations

    large quantities of people

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

    The neuron specializes in spotting numerals and numbers—particularly multi‐digit figures and numeric measures (years, counts, large quantities).

    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
    ↵↵
    -2.00
    0
    -1.83
    1
    -1.68
     entirety
    -1.63
     Provides
    -1.52
        
    -1.49
     savag
    -1.45
     Typical
    -1.43
    possibly
    -1.40
     acoged
    -1.40
    POSITIVE LOGITS
    瑧
    1.97
     of
    1.87
     regler
    1.84
     ditem
    1.80
     būs
    1.76
     himſelf
    1.74
     kommentar
    1.68
     große
    1.65
     miliar
    1.63
     vienas
    1.63
    Activations Density 0.048%

    No Known Activations