© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 59079
    Prev
    Next
    INDEX
    Explanations

    Dutch and German articles

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

    The neuron is detecting common German and Dutch article tokens (like “de,” “Der,” “einen,” etc.).

    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
     Schema
    -0.84
    stadium
    -0.83
    sió
    -0.81
    schema
    -0.81
    Thema
    -0.79
    idium
    -0.75
    crata
    -0.75
     schema
    -0.75
    avanje
    -0.75
    íram
    -0.75
    POSITIVE LOGITS
    dibly
    0.83
    Cette
    0.82
    entreprise
    0.79
     États
    0.79
    RRect
    0.77
    tice
    0.77
     Texto
    0.76
     §§
    0.75
    marke
    0.74
    にかけて
    0.74
    Activations Density 0.006%

    No Known Activations