© 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. 16004
    Prev
    Next
    INDEX
    Explanations

    process, position, public, cancellation, disk

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

    The neuron selectively activates on numeric tokens—especially numbers with decimals.

    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
    ТИ
    0.79
    مع
    0.78
    嶈
    0.77
    Ци
    0.74
    Ле
    0.73
    ита
    0.72
    Ди
    0.71
    Ни
    0.71
    Щ
    0.71
    ವರಿ
    0.70
    POSITIVE LOGITS
    aster
    1.02
    is
    0.87
    ist
    0.86
    xj
    0.80
    érm
    0.80
    ont
    0.79
    pst
    0.79
    om
    0.79
    orton
    0.76
    prices
    0.74
    Activations Density 0.001%

    No Known Activations