© 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. 106518
    Prev
    Next
    INDEX
    Explanations

    Mayan glyphs and history

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

    The neuron activates on numeric date and year tokens (e.g. “2012,” “date,” “end of the age”), i.e. it’s looking for date- or year-related numbers.

    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
     medieval
    -0.89
     Latina
    -0.88
     African
    -0.81
    Depending
    -0.77
     migrator
    -0.77
     Archangel
    -0.76
     morena
    -0.75
    afterEach
    -0.75
     moreno
    -0.75
     Alger
    -0.74
    POSITIVE LOGITS
     glyph
    1.16
     maize
    1.01
     Maize
    0.97
     Maya
    0.96
     Hier
    0.92
     jaguar
    0.91
    jaguar
    0.84
    Puma
    0.83
    elite
    0.82
    mayan
    0.82
    Activations Density 0.012%

    No Known Activations