© 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. 107473
    Prev
    Next
    INDEX
    Explanations

    code snippets and URLs

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

    The neuron strongly fires on code-related tokens—especially JavaScript filenames, plugin names, and file‐extension markers (e.g. “.js,” “Slides,” “slideshow,” etc.).

    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
     hi
    -0.75
    仲
    -0.73
    cti
    -0.73
     progenitor
    -0.72
    ViewImports
    -0.70
     tri
    -0.70
    ole
    -0.69
    override
    -0.69
    wire
    -0.68
     rinse
    -0.68
    POSITIVE LOGITS
     Bullets
    0.92
     LAY
    0.91
    lapis
    0.89
     procedimentos
    0.87
     Sequo
    0.86
     saffron
    0.82
     Robe
    0.82
     vode
    0.81
     REVOLUTION
    0.80
    ום
    0.79
    Activations Density 0.017%

    No Known Activations