© 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. 10-GEMMASCOPE-RES-131K
    7. 42741
    Prev
    Next
    INDEX
    Explanations

    programming code variable names like id, i, j, end

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

    The neuron consistently activates on 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_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
     your
    -1.74
     is
    -1.67
     to
    -1.67
     reputed
    -1.51
    ations
    -1.39
    𝑃
    -1.38
    譏
    -1.35
    lgari
    -1.34
    と言われて
    -1.34
    囁
    -1.32
    POSITIVE LOGITS
    4
    1.80
    3
    1.45
    Созда
    1.41
    Jalan
    1.38
    2
    1.33
    喔
    1.32
    6
    1.32
    
    1.30
     yaitu
    1.30
     není
    1.29
    Activations Density 0.002%

    No Known Activations