© 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. 34-GEMMASCOPE-RES-131K
    7. 31299
    Prev
    Next
    INDEX
    Explanations

    California places, national parks, wildfires

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

    The neuron strongly activates on proper nouns—especially geographic place names (e.g., county, forest, town names).

    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
    Winnipeg
    -0.81
     Lehrer
    -0.78
     episodio
    -0.72
    DEF
    -0.72
    Lukas
    -0.71
    Senegal
    -0.71
    kates
    -0.69
    驿
    -0.69
     chodzi
    -0.69
     Hudson
    -0.69
    POSITIVE LOGITS
     Shasta
    1.38
     Borno
    1.00
     Chico
    1.00
     Redding
    0.99
     Lassen
    0.88
    Oro
    0.83
    leun
    0.81
     Butte
    0.80
     داع
    0.78
     Adventist
    0.78
    Activations Density 0.018%

    No Known Activations