© 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. 10-GEMMASCOPE-RES-131K
    7. 100453
    Prev
    Next
    INDEX
    Explanations

    URLs and file paths

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

    URL paths and web addresses, particularly the components following domain names like slashes, file extensions, and query parameters.

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

    The neuron activates on parts of web addresses (URLs), identifying tokens like “http://”, domain names, slashes, and query parameters.

    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
     curioso
    -1.38
     всю
    -1.38
     chaleco
    -1.36
     vasto
    -1.34
     curiosa
    -1.33
     прошел
    -1.32
     asas
    -1.31
     fluido
    -1.28
     pantalón
    -1.27
     kaki
    -1.27
    POSITIVE LOGITS
    which
    1.98
     that
    1.85
     που
    1.57
     که
    1.57
     or
    1.43
    that
    1.42
     или
    1.39
    któ
    1.38
     التي
    1.37
    Rozmiar
    1.34
    Activations Density 0.036%

    No Known Activations