© 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. 34-GEMMASCOPE-RES-131K
    7. 93702
    Prev
    Next
    INDEX
    Explanations

    programming language keywords

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

    The neuron fires on programming‐language keywords that control resources or data flow—e.g. defer, delete, new, out, uniform, readonly—i.e. tokens used for memory/resource management and variable qualifiers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     célè
    -1.97
    
    -1.88
     corrom
    -1.84
    
    -1.84
    蛱
    -1.82
     délais
    -1.81
     rése
    -1.80
    쐈
    -1.74
     diriger
    -1.74
     différentes
    -1.73
    POSITIVE LOGITS
     in
    2.52
     at
    2.27
     as
    2.25
     with
    2.20
     during
    2.08
     from
    1.99
     or
    1.91
     without
    1.86
     after
    1.86
     although
    1.82
    Activations Density 0.031%

    No Known Activations