© 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. 68374
    Prev
    Next
    INDEX
    Explanations

    court cases citations

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

    The neuron is specialized to pick out the numeric volume and page‐number portions of legal case citations (e.g. the numbers in parentheses and reporter page references).

    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
     می
    -1.07
     |
    -1.05
    --
    -1.05
     LA
    -1.01
     --
    -0.99
    SC
    -0.96
    ,
    -0.94
    I
    -0.94
    Moon
    -0.94
    October
    -0.94
    POSITIVE LOGITS
    کفش
    1.16
     prét
    1.16
    Katalog
    1.12
     efekty
    1.10
     bebés
    1.10
     rafraî
    1.09
     which
    1.09
     trône
    1.06
     botines
    1.05
     médicament
    1.03
    Activations Density 0.217%

    No Known Activations