© 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. 22-GEMMASCOPE-RES-131K
    4. 130632
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numeric tokens—especially multi‐digit numbers such as years, scores, and other quantitative figures.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    subcommittee deals with FDA

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

    descriptions of organizations, entities, or companies - particularly when explaining what they are or their function.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
    Not in Any Lists

    No Comments

    Negative Logits
     and
    -1.16
    Both
    -0.99
    ökning
    -0.93
     with
    -0.90
    心
    -0.89
     скриншот
    -0.86
    \|_{
    -0.82
    Those
    -0.82
    ühl
    -0.77
    That
    -0.77
    POSITIVE LOGITS
    跑了
    1.05
     filós
    1.02
     his
    1.02
     ambienta
    0.98
     require
    0.96
     Ambiental
    0.95
     membutuhkan
    0.94
     colega
    0.94
    ksessa
    0.94
     nutrientes
    0.94
    Activations Density 0.076%

    No Known Activations