© 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. 58888
    Prev
    Next
    INDEX
    Explanations

    introductions and definitions

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

    The neuron is detecting capitalized proper nouns and title‐style phrases (e.g. names, places, and section headings).

    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
     almo
    -1.55
    tament
    -1.52
     stili
    -1.45
    inoco
    -1.43
    撅
    -1.42
    elding
    -1.42
     beserta
    -1.41
    ቕ
    -1.41
     inoltre
    -1.40
     obat
    -1.39
    POSITIVE LOGITS
     if
    1.58
     according
    1.42
     should
    1.40
     what
    1.35
     had
    1.30
     (
    1.27
     one
    1.20
     was
    1.15
     would
    1.14
    <b>
    1.13
    Activations Density 0.274%

    No Known Activations