© 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. 104744
    Prev
    Next
    INDEX
    Explanations

    biographies and scientific classifications

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

    The neuron fires on structured metadata elements—especially numbers (years, measurements) and section‐heading labels (e.g. “Plot,” “Cast,” “Distribution,” taxonomic family endings like “-idae”)—rather than ordinary prose.

    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
     blusa
    -1.20
     superbes
    -1.14
     película
    -1.12
     incroy
    -1.12
    ({
    -1.11
     afbeeldingen
    -1.08
     proposé
    -1.07
     usando
    -1.06
     soyez
    -1.04
     başlay
    -1.04
    POSITIVE LOGITS
     after
    1.34
     will
    1.29
    定期
    1.04
     one
    1.02
     as
    1.00
     on
    0.97
     arī
    0.95
     is
    0.94
     remained
    0.94
     but
    0.94
    Activations Density 0.003%

    No Known Activations