© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 39596
    Prev
    Next
    INDEX
    Explanations

    не- prefix words

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

    The neuron consistently fires on the negative-prefix “ne-/не-” (and its close variants) across different Slavic languages.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     technisch
    -2.27
    邶
    -2.19
    ೣ
    -2.17
    葑
    -1.98
     '
    -1.94
    罍
    -1.91
    狴
    -1.91
    𓇼
    -1.90
    lusconi
    -1.88
    𐄁
    -1.87
    POSITIVE LOGITS
    ’
    2.50
    ’,
    2.48
    ↵
    2.42
    ’，
    2.42
    ‘‘
    2.27
    “,
    2.09
    ：“
    2.05
    ’?
    2.00
    ’.
    1.99
    if
    1.96
    Activations Density 0.013%

    No Known Activations