© 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. 157
    Prev
    Next
    INDEX
    Explanations

    division of nations and states

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

    The neuron consistently activates on numeric tokens—particularly years, dates, and other numbers (including numbers in parentheses)—so it’s detecting date/number expressions.

    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
     ২০১
    -0.88
     平成
    -0.87
     postmodern
    -0.86
     acredita
    -0.84
     descobrir
    -0.83
     intelig
    -0.82
     sabem
    -0.82
    
    -0.81
     dizem
    -0.80
     Harare
    -0.79
    POSITIVE LOGITS
     League
    1.05
     Partition
    1.04
     USSR
    0.97
     ICS
    0.96
     Constitu
    0.92
    Partition
    0.88
    Telegram
    0.88
     telegram
    0.88
    telegram
    0.87
     penicillin
    0.87
    Activations Density 0.037%

    No Known Activations