© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 39791
    Prev
    Next
    INDEX
    Explanations

    URLs and file paths

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

    The neuron fires on web‐address fragments (e.g. protocol tokens like “http”/“https,” slashes “://,” domain names, and path segments), effectively detecting URLs.

    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
     before
    -1.44
     it
    -1.41
     as
    -1.37
     not
    -1.31
     on
    -1.31
     first
    -1.30
     this
    -1.27
     creates
    -1.26
     just
    -1.23
     despite
    -1.19
    POSITIVE LOGITS
     ludicrous
    1.38
     famed
    1.32
     exceptionally
    1.28
    
    1.27
     momentous
    1.27
     insanely
    1.24
     bewildered
    1.23
     wondrous
    1.23
    をご覧
    1.23
     philanth
    1.22
    Activations Density 0.001%

    No Known Activations