© 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. 75371
    Prev
    Next
    INDEX
    Explanations

    medical conditions, events, or proper nouns followed by context

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

    references to important cultural, religious, or historical entities like holidays, diseases, Supreme Court cases, and famous figures.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron detects named entities—proper names of people, places, organizations, events, or specific conditions.

    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
     everything
    -1.19
    everything
    -1.02
     in
    -0.98
     indoors
    -0.90
     themes
    -0.88
     inks
    -0.87
     aislados
    -0.85
     אנחנו
    -0.85
    しかった
    -0.85
    túi
    -0.84
    POSITIVE LOGITS
     felipe
    0.98
    ≳
    0.92
    ₨
    0.92
     eduardo
    0.91
     sexta
    0.91
     ministre
    0.91
    关于
    0.91
    bentar
    0.90
    êtres
    0.90
     mulus
    0.90
    Activations Density 0.212%

    No Known Activations