© 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. 7055
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on quantifying language—numbers, years, statistics, projections, denominations, and similar business‐style metrics.

    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
    非常的
    1.00
    这样一个
    0.95
    很多的
    0.95
    非常に
    0.95
     bzw
    0.93
    னுடைய
    0.93
     अथवा
    0.91
     различные
    0.89
     ആണ്
    0.89
     данном
    0.88
    POSITIVE LOGITS
     Tsuk
    0.85
     berharap
    0.84
     Corriere
    0.82
     Preschool
    0.80
     Macros
    0.79
     Saras
    0.79
     Kha
    0.78
     Tomás
    0.78
     biar
    0.78
     ไหน
    0.77
    Activations Density 0.820%

    No Known Activations