© 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. 23660
    Prev
    Next
    INDEX
    Explanations

    XML schema definitions

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

    The neuron spikes on URL and file‐path punctuation (e.g. “http”, “https”, “. ” separators) – in other words, it’s looking for parts of web addresses and dotted paths.

    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
     SIB
    -0.71
     KK
    -0.68
    weeb
    -0.68
    île
    -0.65
     SAT
    -0.65
    是非常
    -0.65
     DMR
    -0.65
     ats
    -0.64
    geometry
    -0.63
     sân
    -0.63
    POSITIVE LOGITS
    𝗷
    0.77
    𝟙
    0.72
     nourriture
    0.70
     acceptance
    0.68
     MaterialApp
    0.68
    linker
    0.67
     Ewing
    0.66
    machus
    0.66
     recette
    0.66
    Experiments
    0.65
    Activations Density 0.060%

    No Known Activations