© 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. 10058
    Prev
    Next
    INDEX
    Explanations

    eastern, japanese, chinese origins

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

    This neuron strongly activates on auxiliary forms of the verb “be” (e.g. is, be, already [as in “already cooked”]) and related function words.

    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
    イギリス
    0.75
    イタリア
    0.74
     イタリア
    0.74
     Algarve
    0.73
    虻
    0.73
    スペイン
    0.69
     irland
    0.66
    🥐
    0.65
     Huguen
    0.64
    イング
    0.64
    POSITIVE LOGITS
     Chinese
    2.48
     китай
    2.45
    Chinese
    2.38
     chinese
    2.22
     চীনা
    2.19
     Confucian
    2.17
     중국
    2.16
     Китай
    2.13
     Cantonese
    2.08
     cinese
    2.08
    Activations Density 0.888%

    No Known Activations