© 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. 54190
    Prev
    Next
    INDEX
    Explanations

    legal document identifiers

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

    This neuron activates on numerical sequences, especially digits used in legal citations (e.g. volume, page, year, and case numbers).

    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
     дописавши
    -0.85
     shouting
    -0.83
     boneca
    -0.82
     already
    -0.81
     entire
    -0.80
     молоде
    -0.77
    erapeutics
    -0.75
     management
    -0.75
     even
    -0.75
     franqu
    -0.74
    POSITIVE LOGITS
     после
    0.88
    sprzed
    0.86
     juillet
    0.86
     Þ
    0.85
     leuke
    0.85
    POINT
    0.83
    应该
    0.83
     بای
    0.82
     سال
    0.81
     czerwca
    0.81
    Activations Density 0.132%

    No Known Activations