© 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. 32731
    Prev
    Next
    INDEX
    Explanations

    foreign language words

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

    The neuron is selectively activating on common-content nouns (including gerund forms) that name key objects or actions in the text.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    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
     and
    -1.16
    also
    -1.15
    on
    -1.15
    just
    -1.07
    stande
    -1.05
    y
    -1.05
    is
    -0.99
    see
    -0.99
    库存
    -0.99
    at
    -0.98
    POSITIVE LOGITS
     supuestamente
    1.37
     ayudará
    1.37
     Vergnügen
    1.35
     Sünde
    1.34
     nació
    1.34
     continúa
    1.32
    Belgique
    1.32
     ejerc
    1.32
     Persön
    1.30
     elétrico
    1.29
    Activations Density 0.145%

    No Known Activations