© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 5256
    Prev
    Next
    INDEX
    Explanations

    Japanese and Korean grammatical endings

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

    The neuron is detecting non-Latin (East Asian) script characters—e.g. Japanese kanji/kana and Korean hangul.

    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
    Picked
    0.82
    generating
    0.81
    Increasing
    0.79
    侑
    0.79
    的值
    0.75
    Mention
    0.75
    Hydrochloride
    0.75
    Graphs
    0.74
    牘
    0.74
    あります
    0.74
    POSITIVE LOGITS
     것이
    1.59
     것은
    1.48
     것으로
    1.48
     것을
    1.41
     것에
    1.37
     것도
    1.36
     것
    1.30
     곳
    1.28
    것
    1.28
     şey
    1.25
    Activations Density 0.016%

    No Known Activations