© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 1065
    Prev
    Next
    INDEX
    Explanations

    WebAssembly, wasm, ems, wasi

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

    The neuron fires strongly on code-style tokens—especially uppercase acronyms, version numbers, numeric strings, and other technical identifiers (e.g. “WASM,” “2017,” file paths, URLs).

    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
    ėse
    -0.73
     jeudi
    -0.69
     mało
    -0.68
     ocupa
    -0.67
    Coal
    -0.67
    erequisites
    -0.67
     Farah
    -0.67
     comorbidities
    -0.67
    Shades
    -0.67
    強い
    -0.66
    POSITIVE LOGITS
     WAS
    1.47
     wasm
    1.28
    ems
    1.23
     Was
    1.17
     Ems
    1.15
    wasm
    1.10
     Module
    1.02
    Was
    1.01
     ems
    1.00
    wasi
    0.91
    Activations Density 0.057%

    No Known Activations