© 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. 22-GEMMASCOPE-RES-131K
    7. 65818
    Prev
    Next
    INDEX
    Explanations

    HTML tags, UI elements, biological classifications

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

    HTML/XML opening tags for structural elements like div, p, or view.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron detects markup‐style syntax, especially HTML/XML tags and their surrounding angle brackets, slashes, and attribute markers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    Adds
    -0.94
    去年
    -0.87
     new
    -0.85
    yship
    -0.85
     dangere
    -0.84
     prision
    -0.84
    WATCH
    -0.83
    GFX
    -0.83
     Wartawan
    -0.81
    Preço
    -0.81
    POSITIVE LOGITS
    ˂
    0.96
    メージ
    0.88
     elektrische
    0.85
     finalize
    0.85
     caption
    0.84
     ingeniero
    0.84
     captions
    0.84
     áng
    0.84
    的想法
    0.84
    0.82
    Activations Density 0.023%

    No Known Activations