© 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. 10-GEMMASCOPE-RES-131K
    7. 43416
    Prev
    Next
    INDEX
    Explanations

    register followed by code elements

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

    The neuron fires on code occurrences of the word “register” (in any capitalization) indicating register calls or declarations.

    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
    埽
    -3.48
     we
    -3.20
    🫅
    -2.94
     marvelous
    -2.81
     wonderfully
    -2.78
    ي
    -2.77
    蟶
    -2.72
     dismal
    -2.66
     magnificent
    -2.63
     fabulous
    -2.61
    POSITIVE LOGITS
    ”;
    3.59
    ”:
    3.39
    ”?
    3.34
    !”
    3.33
    ”)
    3.17
    Donativos
    3.16
    ...”
    3.16
    .”
    3.11
    ;”
    3.09
     Surprisingly
    3.06
    Activations Density 0.018%

    No Known Activations