© 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. 39526
    Prev
    Next
    INDEX
    Explanations

    code structure or markup

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

    The neuron fires on closing delimiters in code—especially closing braces (“}”) and self-closing tag endings (“/>”).

    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
    ˧
    -0.94
     by
    -0.93
     uiteindelijk
    -0.91
     parameters
    -0.89
     corriendo
    -0.89
     ObjectParameter
    -0.85
    ִּ
    -0.84
     each
    -0.83
     Illinois
    -0.83
    mässig
    -0.82
    POSITIVE LOGITS
     topp
    0.99
     snippetHide
    0.89
     fortal
    0.87
     etab
    0.86
    sibilidade
    0.83
     sombr
    0.83
     controles
    0.82
     trou
    0.82
     fej
    0.81
     nei
    0.81
    Activations Density 0.005%

    No Known Activations