© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 127686
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly lights up on formal, factual exposition passages—especially medical or technical descriptions with common function words (e.g. “at,” “in,” “can,” “be,” “contracted,” “affects”).

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408

    all or any

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    Negative Logits
     after
    -2.55
     with
    -2.48
     or
    -2.33
     by
    -2.23
     in
    -2.17
     using
    -2.14
     on
    -2.05
     Some
    -2.05
     it
    -1.98
    Some
    -1.98
    POSITIVE LOGITS
     모든
    3.58
     everything
    3.52
    EVERY
    3.16
     всички
    3.13
     all
    2.95
     تمامی
    2.72
     всі
    2.64
     усі
    2.64
     everywhere
    2.64
     everyone
    2.64
    Activations Density 0.096%

    No Known Activations