© 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. 467
    Prev
    Next
    INDEX
    Explanations

    numbers and URLs

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

    the neuron detects numeric tokens or multi‐digit number sequences (e.g. years, volume/page numbers, citation digits).

    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
    -
    0.27
    '
    0.25
    ’
    0.23
     part
    0.21
    .
    0.21
     and
    0.20
     full
    0.20
    ?
    0.20
     of
    0.20
     is
    0.19
    POSITIVE LOGITS
     इनवर्स
    0.23
     تساعد
    0.22
     चैप्टर्स
    0.21
     โจทย์
    0.21
     يساعد
    0.21
     ヘア
    0.21
     প্রমুখ
    0.21
    riterien
    0.20
     สืบค้น
    0.20
    ₄
    0.20
    Activations Density 0.373%

    No Known Activations