© 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. 22-GEMMASCOPE-RES-131K
    7. 34219
    Prev
    Next
    INDEX
    Explanations

    code commands or actions

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

    The neuron activates on code comment markers and the comment text itself (e.g. “// …”).

    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
    <bos>
    -1.22
    Sådan
    -1.13
    hướng
    -1.13
     parachoque
    -1.10
     campeonato
    -1.08
    Studi
    -1.07
    zaron
    -1.05
    氳
    -1.05
    Televis
    -1.05
    Familie
    -1.05
    POSITIVE LOGITS
     also
    1.29
     estranho
    1.12
    clark
    1.06
    ^{[
    1.00
     May
    0.99
     Naval
    0.99
    becca
    0.96
     analisar
    0.96
     If
    0.96
    huit
    0.96
    Activations Density 0.006%

    No Known Activations