© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 84314
    Prev
    Next
    INDEX
    Explanations

    court case filings

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

    The neuron activates on the all-caps header and boilerplate metadata lines (e.g. court names, dates, docket numbers, copyright/license notices) at the top of legal filings.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    蕷
    -3.09
    ⑉
    -2.78
     With
    -2.56
    ☆☆☆☆
    -2.42
     atau
    -2.39
     burgeoning
    -2.30
     modelleri
    -2.22
    !!!!!
    -2.22
    Despite
    -2.20
    2
    -2.17
    POSITIVE LOGITS
    鸶
    3.61
    ’
    3.38
    雠
    2.64
    
    2.53
    Ꭳ
    2.53
     dê
    2.52
     sê
    2.52
    赑
    2.48
    鸸
    2.48
     was
    2.47
    Activations Density 0.024%

    No Known Activations