© 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. 31-GEMMASCOPE-2-RES-16K
    4. 369
    Prev
    Next
    INDEX
    Explanations

    The neuron fires most strongly on very common English function words (e.g. “in,” “the,” “to,” “of,” “and”).

    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
     매출
    0.41
     법칙
    0.38
    poč
    0.36
    HttpStatus
    0.35
    Precio
    0.35
    人々
    0.34
    邨
    0.34
    TotalPrice
    0.33
     thấm
    0.33
    UnitPrice
    0.33
    POSITIVE LOGITS
     folder
    0.82
    文件夹
    0.75
     file
    0.72
     ~/.
    0.70
     folders
    0.69
     Datei
    0.66
     downloaded
    0.65
     файл
    0.64
     *.
    0.62
    フォルダ
    0.62
    Activations Density 1.371%

    No Known Activations