© 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. 22-GEMMASCOPE-RES-131K
    7. 68880
    Prev
    Next
    INDEX
    Explanations

    specialized programming identifiers

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

    The neuron strongly activates on programming‐language identifiers—especially module/namespace names, import paths, and other code tokens.

    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
    と思われます
    -0.94
     contextLoads
    -0.93
     feltro
    -0.93
     ferramenta
    -0.92
    があるので
    -0.89
    でしょうね
    -0.88
    があり
    -0.88
     possono
    -0.87
    ύπ
    -0.87
    と思っていた
    -0.87
    POSITIVE LOGITS
     ТО
    1.23
     Интернет
    1.07
    十月
    1.04
    _
    1.04
    のおすすめ
    1.03
     تنها
    1.01
     после
    0.98
     Dünya
    0.97
     as
    0.96
     aqui
    0.94
    Activations Density 0.059%

    No Known Activations