© 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. 49945
    Prev
    Next
    INDEX
    Explanations

    Publishers and questions

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

    This neuron detects reactive‐programming tokens in Swift code—especially operators and types like PublishSubject, bind, tap, etc.

    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
     IRB
    -0.82
     joining
    -0.78
     joined
    -0.74
     join
    -0.74
     Ша
    -0.71
    ̲
    -0.69
    żenia
    -0.69
     stopping
    -0.69
    Cosmic
    -0.68
     Les
    -0.68
    POSITIVE LOGITS
    葡萄酒
    0.95
     erased
    0.93
    Cancel
    0.91
    erase
    0.90
    CurrentValue
    0.88
    xase
    0.87
     cancell
    0.87
    Publisher
    0.86
    Publishers
    0.86
     publishers
    0.85
    Activations Density 0.023%

    No Known Activations