© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 49318
    Prev
    Next
    INDEX
    Explanations

    Louisiana

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

    The neuron fires on occurrences of the proper noun “Louisiana,” especially when it appears in legal‐style or document headings.

    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
    mutual
    -0.75
    ProductCategory
    -0.75
     hypotheses
    -0.75
     zaragoza
    -0.73
    傾向
    -0.71
     Nazareth
    -0.71
    acchar
    -0.71
     syste
    -0.70
     akus
    -0.70
    foul
    -0.69
    POSITIVE LOGITS
     Louisiana
    1.05
    Louisiana
    0.96
     Err
    0.79
     Cajun
    0.77
    シュ
    0.74
    عر
    0.73
    ТЬ
    0.71
     Jind
    0.71
    reciate
    0.71
     έν
    0.71
    Activations Density 0.041%

    No Known Activations