© 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. 68177
    Prev
    Next
    INDEX
    Explanations

    code structure and keywords

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

    The neuron activates on code tokens that signal unit‐testing constructs—test block keywords (like “when:”), test function calls (e.g. test, jest.fn), mocking verbs (patch), and framework helpers (supertest, enzyme’s mount/shallow).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    <strong>
    -1.68
     was
    -1.52
     допомогою
    -1.50
    la
    -1.47
    -
    -1.41
    </h3>
    -1.38
    in
    -1.38
    (
    -1.37
    -1.32
     sont
    -1.31
    POSITIVE LOGITS
     推薦
    1.52
    niosek
    1.44
    夢想
    1.44
    Лечение
    1.42
    偁
    1.40
    傚
    1.39
    ┰
    1.34
    一九
    1.33
     dłu
    1.33
    Beberapa
    1.32
    Activations Density 0.002%

    No Known Activations