© 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. 82792
    Prev
    Next
    INDEX
    Explanations

    URL parameters like utm_source and hl

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

    The neuron activates on URL and query‐string components (e.g. “?”, “=”, “&”, domain names, path segments, and parameter tokens) in 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_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
    urgence
    -0.83
     internal
    -0.76
     lorsque
    -0.75
     głów
    -0.75
    gni
    -0.74
    檄
    -0.73
    (!_
    -0.73
     vertra
    -0.72
     replies
    -0.71
    FloatField
    -0.70
    POSITIVE LOGITS
    source
    1.02
    hl
    0.90
    client
    0.81
    maschinen
    0.81
    DQ
    0.78
    Michele
    0.76
     също
    0.76
    SOURCE
    0.76
    Mitglied
    0.75
    瑞士
    0.75
    Activations Density 0.020%

    No Known Activations