© 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. 66339
    Prev
    Next
    INDEX
    Explanations

    programs, industry, or candidacy

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

    The neuron fires on prominent topic-signaling nouns—i.e. the main content words that carry a document’s key subjects.

    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
     a
    -1.40
     also
    -1.16
     überprü
    -1.15
     his
    -1.13
     fellow
    -1.12
     écoulé
    -1.09
     czyli
    -1.09
    RE
    -1.08
     wollte
    -1.05
     antworte
    -1.05
    POSITIVE LOGITS
     hybride
    1.15
     inexist
    1.15
     auteur
    1.14
     anno
    1.13
     HIT
    1.13
     churras
    1.13
    重さ
    1.11
     Dimen
    1.11
     Rusia
    1.10
     divis
    1.09
    Activations Density 0.052%

    No Known Activations