© 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. 24-GEMMASCOPE-2-RES-16K
    7. 366
    Prev
    Next
    INDEX
    Explanations

    This neuron strongly responds to subword tokens that begin with uppercase letters (e.g. proper names or sentence-initial capitals).

    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
    絶対
    0.21
     FIXME
    0.21
    逡
    0.20
    ([
    0.20
    เกิดขึ้น
    0.20
    অর্চনা
    0.20
    朿
    0.20
    Celltype
    0.19
     Reihen
    0.19
    想定
    0.19
    POSITIVE LOGITS
     యొక్క
    0.27
    ítás
    0.26
    é
    0.25
    aların
    0.25
    anwhile
    0.25
    ä
    0.25
    ları
    0.24
    ين
    0.24
    Ꭳ
    0.24
    گاه
    0.23
    Activations Density 7.372%

    No Known Activations