© 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. 22-GEMMASCOPE-RES-131K
    7. 35549
    Prev
    Next
    INDEX
    Explanations

    foreign words and phrases

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

    The neuron detects metadata or byline markers—words that label sources, authorship, or posting info (e.g. “posted,” “from,” department names, site‐source tokens).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    прият
    -1.20
    </table>
    -1.07
    ([^
    -1.01
    )));
    -0.98
    ostat
    -0.95
    });
    -0.95
    Ї
    -0.94
     निम्नलिखित
    -0.93
    UnknownFieldSet
    -0.93
    endsection
    -0.92
    POSITIVE LOGITS
    Variante
    1.21
    förderung
    1.09
     culturali
    1.05
    質感
    1.04
     situazioni
    1.00
     sırt
    0.99
    wurde
    0.99
     échange
    0.99
    Aantal
    0.99
     gravidanza
    0.98
    Activations Density 0.000%

    No Known Activations