© 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. 45391
    Prev
    Next
    INDEX
    Explanations

    img tags in html

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

    The neuron strongly activates on HTML tag delimiters (the “<…>” syntax and related quote‐delimited attributes), i.e. it’s picking out HTML tags.

    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
     links
    -1.03
     link
    -1.01
     linking
    -0.90
    links
    -0.86
     linkage
    -0.81
    網址
    -0.79
     入力
    -0.79
    URLConnection
    -0.79
    LINKS
    -0.79
    Collegamenti
    -0.77
    POSITIVE LOGITS
    VIRON
    0.90
     mūsų
    0.80
     soir
    0.79
    TPM
    0.78
     thumbnail
    0.77
    𝐴
    0.77
    img
    0.77
    JPG
    0.76
    palabra
    0.75
    jpg
    0.74
    Activations Density 0.006%

    No Known Activations