© 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
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 10924
    Prev
    Next
    INDEX
    Explanations

    它总是

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

    This neuron detects Chinese (Han) characters or Chinese-language text segments.

    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
    이고
    0.77
     일본
    0.65
     스
    0.64
    そうで
    0.62
    치
    0.60
     제
    0.59
    、《
    0.59
    《
    0.58
    、“
    0.57
     치
    0.57
    POSITIVE LOGITS
     কখনোই
    1.32
     всегда
    1.28
    并不是
    1.26
    不僅
    1.25
    不仅
    1.24
     always
    1.20
    並沒有
    1.20
    并没有
    1.19
     memang
    1.18
    គឺ
    1.18
    Activations Density 0.044%

    No Known Activations