© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 32061
    Prev
    Next
    INDEX
    Explanations

    Secret Service

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

    The neuron fires on multi‐digit numeric tokens, i.e. occurrences of numbers or statistical figures in the 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_34/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
     族
    -0.73
     утеп
    -0.72
    "{\
    -0.72
     successo
    -0.72
     мило
    -0.71
     Steuern
    -0.71
     onBind
    -0.70
    kning
    -0.69
    atività
    -0.68
     weekend
    -0.67
    POSITIVE LOGITS
     Country
    1.03
    annual
    0.88
    Country
    0.88
     State
    0.87
    čiť
    0.86
    렴
    0.85
     Sponsors
    0.85
     sponsor
    0.83
     countries
    0.83
     ミルク
    0.82
    Activations Density 0.010%

    No Known Activations