© 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. 913
    Prev
    Next
    INDEX
    Explanations

    The neuron detects numeric tokens and quantifiable expressions (e.g. years, percentages, book‐part numbers, and other 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_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
     desloc
    0.62
     पछ
    0.57
     इक्वेशन
    0.56
    残留
    0.56
     estructuras
    0.56
     grasa
    0.55
     수행
    0.55
    씻
    0.53
     valuables
    0.53
     করিতেছি
    0.53
    POSITIVE LOGITS
     출시
    2.31
     launch
    2.20
    発売
    2.20
     launched
    2.14
     debuted
    2.13
     released
    2.09
     release
    2.03
    launched
    2.00
    released
    1.99
     debuts
    1.95
    Activations Density 1.455%

    No Known Activations