© 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. 98217
    Prev
    Next
    INDEX
    Explanations

    generic type parameters

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

    This neuron detects opening bracket tokens in code—especially the “<” used for generics/templates (and similarly “ [ ”).

    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
     until
    -1.36
     September
    -1.16
     at
    -1.12
    ↵↵
    -1.09
     jornada
    -1.06
     but
    -1.05
     an
    -1.05
     before
    -1.04
     p
    -1.02
    </td>
    -1.01
    POSITIVE LOGITS
     patienter
    1.20
     zosta
    1.20
    EVEN
    1.19
     ukry
    1.19
     maniere
    1.16
    Введение
    1.13
     acheter
    1.09
     leſs
    1.09
     juſt
    1.09
    킁
    1.08
    Activations Density 0.006%

    No Known Activations