© 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. 24-GEMMASCOPE-2-RES-16K
    7. 6761
    Prev
    Next
    INDEX
    Explanations

    conjunctions in Slavic languages

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

    The neuron fires on non-English, diacritics-laden tokens—i.e. foreign-language words containing accented or special characters.

    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_24_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
    ହି
    0.80
    厛
    0.77
     Englewood
    0.75
    andidates
    0.75
     SupCt
    0.75
    akumar
    0.75
    amine
    0.74
     प्रदर्शित
    0.74
    LEY
    0.73
    sley
    0.73
    POSITIVE LOGITS
     и
    1.52
     و
    1.37
     και
    1.37
     এবং
    1.21
     καὶ
    1.20
     और
    1.18
    ،
    1.18
     և
    1.17
    และ
    1.16
     và
    1.16
    Activations Density 0.002%

    No Known Activations