© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 11401
    Prev
    Next
    INDEX
    Explanations

    Queensland, polyps, name

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

    scientific and technical terms or specific named entities.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron strongly fires on single tokens that are domain‐specific “names” or codes—e.g. scientific Latin binomials, gene or protein IDs, mine names, numbers/identifiers—i.e. proper nouns and alphanumeric technical terms.

    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
    </em>
    -1.56
    /*!
    -1.48
    甞
    -1.47
     tivi
    -1.45
     luy
    -1.44
    ك
    -1.44
    ßig
    -1.41
    Я
    -1.39
    ति
    -1.38
    ম
    -1.38
    POSITIVE LOGITS
     by
    2.06
     from
    1.98
     this
    1.88
     at
    1.73
     to
    1.70
    9
    1.59
     or
    1.55
     with
    1.54
     in
    1.52
     which
    1.51
    Activations Density 0.193%

    No Known Activations