© 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. 22-GEMMASCOPE-RES-131K
    7. 28602
    Prev
    Next
    INDEX
    Explanations

    SQL `INSERT INTO` with country codes

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

    This neuron activates strongly on URL components (e.g. “http,” “://,” domain names, “.com,” and slashes), effectively detecting web links.

    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
     in
    -1.13
     reisen
    -0.98
     epitom
    -0.88
     overdo
    -0.88
     seven
    -0.87
     Pertama
    -0.85
     as
    -0.85
     Perlu
    -0.84
     Lingkungan
    -0.84
     subsidies
    -0.83
    POSITIVE LOGITS
     multicolore
    1.17
     conster
    1.13
     Поэтому
    1.02
     recens
    1.01
    .}(
    1.01
    aing
    0.97
     этих
    0.96
     ハウス
    0.95
     USES
    0.94
     vécu
    0.94
    Activations Density 0.453%

    No Known Activations