© 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. 127749
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on substantive “content” words (nouns, verbs, numbers, and other non-stop words) that carry the key technical or quantitative meaning in a sentence.

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

    every all whole

    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
     but
    -1.02
     Wirklichkeit
    -0.96
    ૬
    -0.91
    orologio
    -0.85
     permissible
    -0.84
     using
    -0.84
    Aftermath
    -0.82
    اقرأ
    -0.79
     ermöglich
    -0.79
     Within
    -0.79
    POSITIVE LOGITS
     entire
    1.25
     full
    1.13
     all
    1.09
     volledige
    1.08
     volledig
    1.06
    potentially
    1.05
     each
    1.05
    nicy
    1.03
     puisque
    1.02
     EACH
    1.00
    Activations Density 0.057%

    No Known Activations