© 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. 28426
    Prev
    Next
    INDEX
    Explanations

    code syntax

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

    The neuron responds to numeric literals (numbers) in code.

    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
     ký
    -1.06
     Menurut
    -1.01
    computa
    -1.00
     adhesivo
    -0.96
    intem
    -0.96
     måde
    -0.96
     fået
    -0.95
    那些
    -0.95
     varme
    -0.93
     histó
    -0.92
    POSITIVE LOGITS
    oczes
    0.91
    EACH
    0.88
     eccl
    0.88
    様々な
    0.87
     analytically
    0.86
     என்ப
    0.81
    =";
    0.81
     overcome
    0.80
     ecclesias
    0.80
     providing
    0.79
    Activations Density 0.007%

    No Known Activations