© 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. 73817
    Prev
    Next
    INDEX
    Explanations

    ведет, водит, выводит, вода

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

    This neuron detects Russian inflections of the verb “выводить/вывести” (to output or display).

    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
    O
    -2.72
    "
    -2.64
    4
    -2.55
    }
    -2.45
    3
    -2.44
    X
    -2.39
    ization
    -2.39
    6
    -2.28
     expliqué
    -2.23
    {
    -2.22
    POSITIVE LOGITS
    荖
    2.19
    ！！”
    2.05
    哢
    2.05
    agosto
    2.05
    '],
    2.03
    étricas
    2.02
     What
    1.93
     and
    1.93
     bmi
    1.93
    焄
    1.91
    Activations Density 0.003%

    No Known Activations