© 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. 130565
    Prev
    Next
    INDEX
    Explanations

    The neuron detects numerical data tokens (quantities, counts, and measurements) in the text.

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

    from

    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
    Not in Any Lists

    No Comments

    Negative Logits
     in
    -1.38
     on
    -1.05
     will
    -1.05
    ご紹介
    -1.04
    return
    -1.03
     data
    -1.03
     на
    -1.00
    CastException
    -1.00
     using
    -1.00
     в
    -0.98
    POSITIVE LOGITS
     from
    1.96
     koment
    1.08
     pescoço
    1.07
     Estoy
    1.05
    Может
    1.03
     Siempre
    1.02
    することができます
    1.02
     provenant
    1.02
     Hvis
    1.01
     থেকে
    1.01
    Activations Density 0.129%

    No Known Activations