© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 87172
    Prev
    Next
    INDEX
    Explanations

    html head and meta tags

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

    The neuron strongly responds to HTML markup tags (i.e. the angle‐bracketed tag names like `<html>`, `<head>`, `<meta>`, 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
    doise
    -0.88
     ngữ
    -0.82
    servo
    -0.82
    期刊论文
    -0.82
     Languages
    -0.82
    语言
    -0.81
    ácií
    -0.81
     language
    -0.81
    Europe
    -0.80
    }}^{
    -0.80
    POSITIVE LOGITS
    ヘッド
    0.92
    Bibliografía
    0.90
    ucc
    0.84
     gna
    0.79
     Irrit
    0.78
     nib
    0.77
    Irrit
    0.76
     ameste
    0.75
     dam
    0.74
    อีก
    0.73
    Activations Density 0.036%

    No Known Activations