© 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. 66447
    Prev
    Next
    INDEX
    Explanations

    C++ code keywords

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

    This neuron fires on programming identifiers or symbols that contain underscores (e.g. macro names, generated symbols, or library identifiers).

    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
     all
    -2.05
     by
    -1.81
     in
    -1.81
     with
    -1.80
     to
    -1.80
     which
    -1.63
     their
    -1.62
     other
    -1.59
     each
    -1.48
     where
    -1.48
    POSITIVE LOGITS
    튿
    1.66
    了下去
    1.45
    Produkte
    1.39
     beliebt
    1.38
    ATION
    1.32
    感じです
    1.30
     trám
    1.30
                                
    1.28
     Эти
    1.27
    èvement
    1.27
    Activations Density 0.003%

    No Known Activations