© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7944
    Prev
    Next
    INDEX
    Explanations

    generation/decade/era

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

    The neuron strongly activates on timeline markers—years, dates, and other numeric time references in historical or chronologically structured text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    酋
    0.58
     irreducible
    0.57
     реаль
    0.55
     irregular
    0.54
     admiral
    0.53
     दुर्ग
    0.52
     ancestral
    0.52
    损害
    0.51
    苍
    0.51
     infinitely
    0.51
    POSITIVE LOGITS
     decade
    1.88
     دهه
    1.88
     década
    1.75
    decade
    1.73
     시대
    1.70
     era
    1.63
    Generation
    1.62
     generation
    1.60
    年代
    1.59
     Generation
    1.59
    Activations Density 0.396%

    No Known Activations