© 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. 8362
    Prev
    Next
    INDEX
    Explanations

    and informational

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

    This neuron flags prominent content‐bearing nouns (e.g. “Quality,” “Gravitation,” “Genealogy,” “War,” “Power,” etc.) when they appear as key terms in titles or feature listings.

    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
     này
    0.72
     enfim
    0.70
    ங்கு
    0.65
    欢
    0.63
    ême
    0.62
     sciences
    0.62
     tehát
    0.61
     这是
    0.60
     действует
    0.60
    詳
    0.60
    POSITIVE LOGITS
    ndash
    0.82
     its
    0.81
     Its
    0.80
    Its
    0.77
     whatnot
    0.75
    ंगाबाद
    0.74
    Beyond
    0.73
    the
    0.72
    beyond
    0.72
    ERSON
    0.70
    Activations Density 0.109%

    No Known Activations