© 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. 128793
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on programming-language declaration tokens—things like namespace, class, import/use, and other structural keywords that define modules or types.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    test class names

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    теристики
    -0.91
    UAV
    -0.90
     novelas
    -0.89
    ipeline
    -0.87
    ſtre
    -0.84
    ドール
    -0.83
    chemise
    -0.83
    вня
    -0.82
    guro
    -0.82
    <blockquote>
    -0.81
    POSITIVE LOGITS
     test
    1.52
     tests
    1.30
    ServiceTest
    1.14
    test
    1.05
     getestet
    0.97
    RANDOM
    0.91
     tested
    0.91
     Test
    0.91
     testing
    0.90
     \
    0.89
    Activations Density 0.066%

    No Known Activations