© 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. 65776
    Prev
    Next
    INDEX
    Explanations

    Michigan locations and legal citations

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

    The neuron primarily detects legal‐document metadata—strings of uppercase letters and digits such as docket numbers, court identifiers, and other citation markers.

    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
     Wrocław
    -0.79
     ELECTRONIC
    -0.77
    में
    -0.74
    φέρ
    -0.73
    ATHY
    -0.69
    sembles
    -0.67
    inb
    -0.67
    Dear
    -0.67
     androgen
    -0.66
     insufficiency
    -0.65
    POSITIVE LOGITS
     Traverse
    0.95
    Traverse
    0.86
     Antrim
    0.77
    KBS
    0.75
    Чтоб
    0.73
     versión
    0.73
    合
    0.72
    ":[
    0.71
     beneficiaries
    0.71
     Northern
    0.71
    Activations Density 0.018%

    No Known Activations