© 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. 10-GEMMASCOPE-RES-131K
    7. 9539
    Prev
    Next
    INDEX
    Explanations

    programming, science, and specific names

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

    The neuron is detecting named entities or specialized proper nouns (e.g. library/package names, technical class identifiers, person or place names).

    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
    Ⓡ
    -1.42
    ERIE
    -1.38
    Ingrédients
    -1.37
     行き
    -1.31
    şey
    -1.30
    Beaucoup
    -1.29
    ža
    -1.29
    uteurs
    -1.28
    лама
    -1.27
     genç
    -1.26
    POSITIVE LOGITS
    baomidou
    1.73
     that
    1.44
     damaligen
    1.41
     such
    1.39
     是否
    1.31
     Gastgeber
    1.31
    scribers
    1.30
    empio
    1.30
     others
    1.30
     who
    1.28
    Activations Density 0.152%

    No Known Activations