© 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. 259
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on numeric tokens—especially years, dates, quantities, and other large numbers.

    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
     a
    0.36
     the
    0.35
    спользу
    0.30
    定义
    0.30
     an
    0.30
    MBOLS
    0.30
     সর্বদা
    0.29
     lazy
    0.29
    َد
    0.28
    нуть
    0.28
    POSITIVE LOGITS
    .,
    0.44
    .;
    0.44
     όπου
    0.42
     والذي
    0.42
     nhưng
    0.41
    ®.
    0.41
    ؛
    0.41
    .(
    0.39
     whereby
    0.39
    。(
    0.39
    Activations Density 10.825%

    No Known Activations