© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 117978
    Prev
    Next
    INDEX
    Explanations

    plaintiff in legal cases

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

    The neuron selectively activates on formal legal‐document headers and citations—especially party role labels (e.g. “Plaintiff,” “Appellant,” “Appellee”) and case citation details.

    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
     monster
    -0.69
    vmax
    -0.69
     regenerated
    -0.68
    Buen
    -0.66
    bie
    -0.66
     Amsterdam
    -0.66
    specialchars
    -0.65
     chances
    -0.65
     Щ
    -0.65
     Mak
    -0.64
    POSITIVE LOGITS
    ργα
    0.73
    Hvor
    0.73
     아이
    0.72
     Alibaba
    0.71
     Faire
    0.71
    Aufbau
    0.69
    Xe
    0.68
    첸
    0.67
    Pren
    0.67
    profil
    0.66
    Activations Density 0.082%

    No Known Activations