© 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. 8478
    Prev
    Next
    INDEX
    Explanations

    interrogative words in multiple languages

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

    This neuron activates on question words (e.g. “Quel/Quelles” in French or “Какие/Каких” in Russian).

    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
    t
    -2.95
    In
    -2.69
    i
    -2.56
    h
    -2.42
     and
    -2.41
    *
    -2.34
    }
    -2.34
    in
    -2.28
    {
    -2.23
    p
    -2.22
    POSITIVE LOGITS
    utterstock
    3.13
    橚
    2.88
    だけです
    2.78
    좆
    2.61
    泚
    2.59
     alguns
    2.58
    
    2.53
    краси
    2.48
    稣
    2.47
    湏
    2.42
    Activations Density 0.002%

    No Known Activations