© 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. 10-GEMMASCOPE-RES-131K
    4. 104276
    Prev
    Next
    INDEX
    Explanations

    HTML charset UTF-8

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

    the character encoding declaration in HTML meta tags (specifically "charset=UTF-8").

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron specializes in spotting HTML `<meta charset=…>` tags that declare the document’s character encoding.

    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
     podían
    -1.51
     concluyó
    -1.48
     dubbed
    -1.42
    ’,
    -1.42
    _
    -1.33
     должны
    -1.32
    使用的
    -1.31
     from
    -1.29
     unterstüt
    -1.28
     resultó
    -1.27
    POSITIVE LOGITS
     هناك
    1.49
    assertRaises
    1.49
    我没有
    1.47
    꽤
    1.46
    metheus
    1.42
     där
    1.41
     欣赏
    1.40
    VIATIONS
    1.38
    chées
    1.34
    CEEDED
    1.34
    Activations Density 0.020%

    No Known Activations