© 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. 70656
    Prev
    Next
    INDEX
    Explanations

    Category names and namespaces

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

    The neuron responds to high-level structural or technical labels—such as code directives (“using”) and figure references (“FIG”)—in a document.

    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
    向了
    -0.98
     tunik
    -0.96
     sierpnia
    -0.95
    ه‌ای
    -0.94
     akum
    -0.93
    issement
    -0.93
    Lover
    -0.93
    TOE
    -0.90
    áis
    -0.90
    ihnachts
    -0.90
    POSITIVE LOGITS
     
    0.99
    2
    0.94
     في
    0.91
    Among
    0.90
    m
    0.90
    1
    0.89
     medida
    0.87
     suitable
    0.87
    Along
    0.86
    You
    0.86
    Activations Density 0.009%

    No Known Activations