© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 13237
    Prev
    Next
    INDEX
    Explanations

    domain names and urls

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

    The neuron is detecting URL domain tokens—especially hyphenated domain segments and numeric labels commonly found in web addresses.

    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
    Also
    -1.99
     that
    -1.63
    Также
    -1.62
    ());
    -1.57
     також
    -1.55
    ᱾
    -1.55
    También
    -1.54
     также
    -1.54
     */
    -1.48
     навіть
    -1.47
    POSITIVE LOGITS
    逞
    1.40
     it
    1.38
      
    1.23
    cursions
    1.20
     totalmente
    1.20
     who
    1.19
    幼稚
    1.16
    returned
    1.16
    上来
    1.16
     completamente
    1.14
    Activations Density 0.054%

    No Known Activations