© 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. 102974
    Prev
    Next
    INDEX
    Explanations

    URL rewriting and address bar

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

    URLs and web addresses, particularly in technical contexts involving URL rewriting, redirects, and web development.

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

    The neuron fires on mentions of URLs or web addresses (e.g. “URL,” “url,” “http://…,” “.com”) and related configuration terms for URL rewriting.

    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
     mesmas
    -1.37
    峮
    -1.35
    这些
    -1.35
    </h5>
    -1.27
    <bos>
    -1.27
    ↵↵↵↵↵↵↵↵↵↵↵↵
    -1.26
     notamment
    -1.26
    ↵↵↵↵↵↵↵↵↵↵
    -1.25
    ного
    -1.24
     fleuve
    -1.24
    POSITIVE LOGITS
    予定です
    1.32
     the
    1.30
     who
    1.27
     this
    1.27
    念头
    1.23
     Иванович
    1.23
    lo
    1.19
     Imre
    1.19
     estaremos
    1.18
     that
    1.16
    Activations Density 0.098%

    No Known Activations