© 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. 34-GEMMASCOPE-RES-131K
    4. 87171
    Prev
    Next
    INDEX
    Explanations

    products and art

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

    This neuron detects promotional or sales-pitch phrasing—phrases offering recommendations or calls to action (e.g. “If you need to find… we have compiled an excellent selection for you”).

    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
    妻子
    -0.99
     tjän
    -0.89
     strange
    -0.84
     from
    -0.83
     two
    -0.81
     at
    -0.80
     Erben
    -0.80
     four
    -0.79
     three
    -0.78
     free
    -0.78
    POSITIVE LOGITS
    Amazon
    1.20
     buyer
    1.18
     buyers
    1.17
     amazon
    1.09
    amazon
    1.06
     Amazon
    1.06
     product
    1.05
     Buyers
    1.02
     Buyer
    0.96
     użytk
    0.95
    Activations Density 0.006%

    No Known Activations