© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 59736
    Prev
    Next
    INDEX
    Explanations

    запроса or задать

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

    The neuron activates strongly on Cyrillic fragments of the Russian word “запрос” (request/query) in its various inflected forms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    って思
    -0.86
    でしょうか
    -0.81
     renuncia
    -0.81
    って感じ
    -0.80
     acidente
    -0.77
    などを
    -0.76
    قاء
    -0.75
     atve
    -0.75
     ases
    -0.75
     analis
    -0.75
    POSITIVE LOGITS
    arias
    0.99
    protoc
    0.85
    @[+][
    0.85
    IndentedString
    0.84
    entlig
    0.84
     frosted
    0.84
    PostBack
    0.83
    RotationOrder
    0.81
    綜合報導
    0.81
    localctx
    0.81
    Activations Density 0.005%

    No Known Activations