© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 117845
    Prev
    Next
    INDEX
    Explanations

    mason, masonry, lodge

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

    The neuron consistently fires on capitalized tokens and proper‐name or title words (e.g. names of lodges, societies, technical headings, or building names).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    lotta
    -0.76
    нился
    -0.74
     *=
    -0.73
     Dune
    -0.70
     Chapter
    -0.69
    chapter
    -0.69
    museum
    -0.68
    īb
    -0.68
     geen
    -0.67
    הגב
    -0.67
    POSITIVE LOGITS
    Mason
    2.31
     Mason
    2.13
     brick
    2.06
     mason
    2.03
     masonry
    1.93
    mason
    1.90
     MASON
    1.80
     Masonry
    1.80
    Brick
    1.70
     Masonic
    1.66
    Activations Density 0.026%

    No Known Activations