© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 43985
    Prev
    Next
    INDEX
    Explanations

    foreign words and phrases

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

    The neuron fires on the main “topic” tokens of a passage—proper names, key nouns, version numbers, code‐identifier segments, or other focal terms that carry the document’s subject.

    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
     as
    -1.94
     to
    -1.84
     subsequently
    -1.77
     which
    -1.77
     then
    -1.70
     of
    -1.64
     still
    -1.50
    );
    -1.48
     another
    -1.44
    ="
    -1.43
    POSITIVE LOGITS
     sağlar
    1.53
     aislados
    1.52
     posibilidades
    1.38
     aislada
    1.35
     recientes
    1.31
     ayakkab
    1.30
     necesarios
    1.30
    分に
    1.27
    leyeb
    1.27
     cercanos
    1.27
    Activations Density 0.088%

    No Known Activations