© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 104502
    Prev
    Next
    INDEX
    Explanations

    programming keywords and imports

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

    This neuron detects top-level programming directives and declarations (import/package/using/func/require/postulate/public keywords) across different languages.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    ですが
    -1.48
    加減
    -1.29
    などが
    -1.28
    があります
    -1.24
    などは
    -1.22
    かもしれませんが
    -1.17
    比较好
    -1.15
    って思
    -1.14
    と思われます
    -1.11
     모든
    -1.08
    POSITIVE LOGITS
     this
    1.59
    えっ
    1.51
    ܜ
    1.48
    際には
    1.32
    ')")
    1.31
    ごめん
    1.29
    はこんな感じ
    1.23
    ナメント
    1.23
    metry
    1.22
    明日は
    1.22
    Activations Density 0.003%

    No Known Activations