© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7267
    Prev
    Next
    INDEX
    Explanations

    numbers and hotels

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

    This neuron activates on numeric tokens—dates, phone numbers, chapter/verse references, counts, and similar numbers in the text.

    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
     lipoproteins
    0.55
     elves
    0.53
    களில்
    0.50
     வே
    0.50
     darf
    0.50
    ዓ
    0.50
    ロット
    0.49
     ле
    0.47
    ➵
    0.47
     scrollBody
    0.47
    POSITIVE LOGITS
    再
    0.49
    প্রায়
    0.49
     खासकर
    0.48
    fatal
    0.48
     stands
    0.47
     నాలుగు
    0.47
    един
    0.46
     সাজ
    0.46
     re
    0.46
     iniciativas
    0.45
    Activations Density 0.215%

    No Known Activations