© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 71477
    Prev
    Next
    INDEX
    Explanations

    code block structuresThis neuron detects code structures like if/else, class implementations, function definitions, and HTML elements within code snippets

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    neling
    -1.44
     možno
    -1.32
     problemu
    -1.20
     kdy
    -1.20
    つけた
    -1.18
     wers
    -1.17
     ilustracja
    -1.16
    掉的
    -1.15
     mogli
    -1.14
     =
    -1.12
    POSITIVE LOGITS
     this
    1.48
     his
    1.27
    ですが
    1.21
     on
    1.19
     sebuah
    1.16
     from
    1.16
    lahan
    1.14
     yourself
    1.13
    新华
    1.12
     この
    1.10
    Activations Density 0.311%

    No Known Activations