© 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. 71309
    Prev
    Next
    INDEX
    Explanations

    language meaning

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

    This neuron detects formal technical or legal discourse markers and jargon (e.g. “herein,” “below,” “industry professionals”) characteristic of patent-style writing.

    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
     футболка
    -0.93
     оригинал
    -0.90
     to
    -0.90
    houettes
    -0.87
    loyment
    -0.87
    чью
    -0.86
     ideolog
    -0.85
     dátum
    -0.85
     pazi
    -0.84
    jadwal
    -0.84
    POSITIVE LOGITS
     meaning
    1.05
     but
    1.02
     because
    0.97
    omenclature
    0.93
    endiendo
    0.89
     líquida
    0.88
    ears
    0.88
    When
    0.86
     отличие
    0.86
    景点
    0.85
    Activations Density 0.077%

    No Known Activations