© 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. 22-GEMMASCOPE-RES-131K
    7. 130612
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on the “19” (and adjoining digits) of U.S. court citation dates, i.e. four-digit years in the 1900s.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    9 followed by 8 or 7

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     for
    -1.30
     so
    -1.23
     it
    -1.05
     any
    -0.99
    
    -0.97
    DeviceType
    -0.96
     one
    -0.95
     now
    -0.94
     if
    -0.94
     as
    -0.89
    POSITIVE LOGITS
    ícias
    1.11
    sion
    1.05
    want
    1.05
     myriad
    1.02
     twos
    1.01
     revamped
    0.99
    zeitig
    0.96
     ravaged
    0.96
    five
    0.95
     joyful
    0.94
    Activations Density 0.046%

    No Known Activations