© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 43673
    Prev
    Next
    INDEX
    Explanations

    column labels and descriptive terms

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

    This neuron detects table formatting delimiters—specifically the vertical bar “|” characters used to mark columns in text‐based or Markdown‐style tables.

    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
     enjeux
    -1.50
     émissions
    -1.35
     rempliss
    -1.34
     campagnes
    -1.32
    Souls
    -1.30
     épreuves
    -1.29
     publicités
    -1.28
     peuples
    -1.27
     histoires
    -1.27
    Muffin
    -1.26
    POSITIVE LOGITS
     that
    2.14
     what
    1.52
     groups
    1.51
     different
    1.48
     which
    1.48
     those
    1.35
     such
    1.34
     where
    1.34
     these
    1.32
     how
    1.31
    Activations Density 0.033%

    No Known Activations