© 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. 34-GEMMASCOPE-RES-131K
    7. 38217
    Prev
    Next
    INDEX
    Explanations

    law firm names

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

    The neuron flags capitalized names of people, law firms, and other legal entities (e.g. party and counsel names) in court‐document text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    isie
    -0.86
    mortem
    -0.76
    Ɵ
    -0.75
     wär
    -0.74
    publicain
    -0.71
    ewon
    -0.71
    ったのは
    -0.70
     chaleur
    -0.69
     رقم
    -0.69
    Reporting
    -0.68
    POSITIVE LOGITS
     attorney
    1.67
     attorneys
    1.62
     firm
    1.55
     Firm
    1.34
     representation
    1.27
     represents
    1.27
     représentation
    1.23
    firm
    1.20
     representing
    1.18
    Attorney
    1.18
    Activations Density 0.244%

    No Known Activations