© 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. 34852
    Prev
    Next
    INDEX
    Explanations

    US regions and associations

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

    The neuron activates on names of U.S. geographic regions defined by compass directions (e.g., “Midwest,” “Southwest,” “Northwest”).

    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
     contradictory
    -0.79
     contrary
    -0.79
    ků
    -0.79
    offrir
    -0.78
     gesuchten
    -0.78
     expectation
    -0.77
     ótima
    -0.74
     genannten
    -0.73
    だが
    -0.72
     Firstly
    -0.71
    POSITIVE LOGITS
    eners
    1.06
     Missouri
    1.03
     Airlines
    1.00
    🌞
    0.99
    tenth
    0.97
    ern
    0.94
     ern
    0.93
     barcel
    0.90
     Guwahati
    0.89
    0.89
    Activations Density 0.013%

    No Known Activations