© 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. 54222
    Prev
    Next
    INDEX
    Explanations

    conjunctions and separators

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

    This neuron responds to terms describing geometric or structural features (shape, base, sides).

    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
    感じが
    -1.21
     zweifel
    -1.20
     Additionally
    -1.12
    ほうが
    -1.11
     This
    -1.09
     verschiedenen
    -1.09
     vorliegenden
    -1.06
     beinahe
    -1.05
     dasselbe
    -1.05
    のでは
    -1.05
    POSITIVE LOGITS
    šia
    1.30
    そして
    1.22
     but
    1.20
    は
    1.13
    certainly
    1.13
     Quintana
    1.12
    ථ
    1.12
    があり
    1.11
    cketed
    1.08
    Spesifikasi
    1.08
    Activations Density 0.003%

    No Known Activations