© 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. 42021
    Prev
    Next
    INDEX
    Explanations

    associated with technical or medical contexts

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

    nouns that represent technical entities or components like databases, threads, classes, cells, or systems.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron is a “noun detector” that lights up on content‐bearing nouns and key topic terms.

    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
     this
    -1.77
    𒊹
    -1.55
    然後
    -1.50
    skrä
    -1.50
     następnie
    -1.49
    حوال
    -1.48
    här
    -1.45
     ekolog
    -1.44
    jaket
    -1.44
     then
    -1.44
    POSITIVE LOGITS
     its
    2.16
    y
    1.65
     if
    1.65
    e
    1.55
    se
    1.55
    f
    1.52
     to
    1.52
     have
    1.50
    a
    1.48
    ti
    1.45
    Activations Density 0.287%

    No Known Activations