© 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. 130651
    Prev
    Next
    INDEX
    Explanations

    The neuron detects references to advertisements—words like “commercials,” “ad,” “advertisements,” or “promos.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    advertising and marketing

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    event
    -1.07
     ЛИ
    -1.07
    écart
    -1.05
     軽量
    -1.00
    žku
    -0.96
     trattato
    -0.94
    ostante
    -0.93
     possibilidades
    -0.93
    porque
    -0.93
     周辺
    -0.93
    POSITIVE LOGITS
     campaign
    1.63
     marketing
    1.50
     advertising
    1.41
     campaigns
    1.26
     ads
    1.22
     hidratante
    1.21
     commercials
    1.21
     sobremesa
    1.19
     menampilkan
    1.19
     produced
    1.13
    Activations Density 0.038%

    No Known Activations