© 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. 84524
    Prev
    Next
    INDEX
    Explanations

    reporting statistics and data

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

    The neuron fires most strongly on capitalized topic‐heading words or key terms (proper nouns and important nouns) that serve as section titles or titles of items.

    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
     will
    -1.02
    Hilsen
    -0.92
    附近
    -0.91
    Netto
    -0.88
    mvh
    -0.88
    TÜ
    -0.85
    ׇ
    -0.85
    時から
    -0.85
    Veh
    -0.85
    工作
    -0.85
    POSITIVE LOGITS
     records
    1.52
     estimate
    1.48
     recorded
    1.38
     estimates
    1.38
    Records
    1.37
     statistics
    1.34
    records
    1.31
     reported
    1.25
    统计
    1.20
    統計
    1.18
    Activations Density 0.135%

    No Known Activations