© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 10889
    Prev
    Next
    INDEX
    Explanations

    topics and categories

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

    The neuron lights up on content-heavy, relatively uncommon words (especially longer, topic-specific nouns and terms).

    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_24_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
     gefähr
    0.61
    대는
    0.59
    桀
    0.59
     Bedürfn
    0.58
     caos
    0.57
     cz
    0.57
    のは
    0.57
     quiere
    0.56
    たち
    0.55
    と思って
    0.54
    POSITIVE LOGITS
     displayed
    0.73
     반영
    0.69
     سبق
    0.63
     evident
    0.62
     disampaikan
    0.62
     occasioned
    0.59
     विशेषताओं
    0.59
    瑄
    0.59
     that
    0.59
     exhibited
    0.59
    Activations Density 0.055%

    No Known Activations