© 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. 84234
    Prev
    Next
    INDEX
    Explanations

    event or state description

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

    The neuron fires on salient content nouns and proper nouns—words that name key entities, places, or domain-specific concepts.

    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
    -1.24
    院校
    -1.10
     when
    -1.09
     by
    -1.08
     because
    -1.02
     näm
    -1.02
     should
    -0.98
     while
    -0.97
     cucharada
    -0.95
     cucharadita
    -0.95
    POSITIVE LOGITS
     being
    1.75
    being
    1.28
     underway
    1.11
     gone
    1.09
     Uniwers
    1.05
     ヴィンテージ
    1.02
     متعلقه
    1.02
     available
    1.00
    lifer
    0.99
    tifier
    0.99
    Activations Density 0.094%

    No Known Activations