© 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. 12929
    Prev
    Next
    INDEX
    Explanations

    code with functions and prediction

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

    The neuron activates on programming‐language syntax—identifying code tokens such as keywords, function definitions, and other code constructs.

    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
     dlaczego
    -1.26
     chrétien
    -1.10
    也算是
    -1.10
     proč
    -1.04
     obrazov
    -1.01
    偎
    -1.00
    lccc
    -0.99
     näin
    -0.97
    ету
    -0.96
     tämän
    -0.95
    POSITIVE LOGITS
    ,
    1.26
    före
    1.04
    </h2>
    1.01
    别人
    1.00
    传输
    0.97
          
    0.96
    变了
    0.95
    .
    0.93
     you
    0.93
         
    0.92
    Activations Density 0.003%

    No Known Activations