© 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. 108587
    Prev
    Next
    INDEX
    Explanations

    code blocks and testing setups

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

    The neuron selectively lights up on unit‐test code constructs—things like test function declarations and test‐framework keywords (e.g. “func Test…”, “it(…)”, “describe(…)”, beforeEach/afterEach, 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
     this
    -0.85
     呢
    -0.82
    beforeAll
    -0.79
     if
    -0.79
     visited
    -0.77
    re
    -0.77
     küche
    -0.77
    Expect
    -0.75
     begy
    -0.75
     first
    -0.74
    POSITIVE LOGITS
    Finally
    1.17
     finally
    0.98
     Finally
    0.96
    ThemeOverlay
    0.96
     finalmente
    0.95
     наконец
    0.94
     finalement
    0.90
     enfin
    0.87
    Bw
    0.87
    Lastly
    0.86
    Activations Density 0.002%

    No Known Activations