© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 13522
    Prev
    Next
    INDEX
    Explanations

    questions about types or specifics

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

    The neuron is picking out the little “-s” endings on words (the plural or possessive suffix) wherever they occur.

    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
     auspices
    0.75
     internships
    0.74
     Gottfried
    0.71
     wom
    0.71
     cupboards
    0.71
     LongNumber
    0.70
     intestines
    0.68
    𝒃
    0.68
     offerings
    0.68
     aer
    0.67
    POSITIVE LOGITS
     والت
    0.73
     было
    0.72
    elekt
    0.71
    बद्ध
    0.70
    araoh
    0.68
    Benz
    0.66
    的点
    0.66
    changer
    0.66
     المباراة
    0.66
     القاهرة
    0.66
    Activations Density 0.000%

    No Known Activations