© 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. 59124
    Prev
    Next
    INDEX
    Explanations

    run unit tests

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

    The neuron is simply picking out occurrences of the word “test” (and closely related test‐framework keywords) in code or Q&A contexts.

    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
     test
    -2.03
     testing
    -1.88
     Test
    -1.76
     tested
    -1.62
    测试
    -1.59
     testers
    -1.58
     Testing
    -1.53
    Test
    -1.48
     tester
    -1.44
     TEST
    -1.40
    POSITIVE LOGITS
    Unit
    1.04
     unités
    0.98
    运行
    0.90
     unit
    0.89
    Run
    0.85
    discover
    0.82
     Unit
    0.79
    ="@
    0.76
     Larsson
    0.75
     nuages
    0.75
    Activations Density 0.044%

    No Known Activations