© 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. 9686
    Prev
    Next
    INDEX
    Explanations

    ebook publishing

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

    The neuron strongly activates on tokens containing or consisting of numbers (e.g. phone‐number and price digits) and other numeric identifiers.

    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
     literal
    0.61
     cutters
    0.61
     jars
    0.59
    注释
    0.58
     reread
    0.58
     translation
    0.56
     鞋
    0.56
     translations
    0.56
     cutter
    0.54
     Epstein
    0.54
    POSITIVE LOGITS
    Amazon
    0.75
     Amazon
    0.73
    amazon
    0.65
     Kindle
    0.64
     amazon
    0.64
    Caucasian
    0.62
    Ƹ
    0.61
    wordpress
    0.61
     zamanda
    0.61
    <unused29>
    0.60
    Activations Density 0.176%

    No Known Activations