© 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. 82387
    Prev
    Next
    INDEX
    Explanations

    essays, dating, Viagra

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

    This neuron fires on high-value single words used as SEO or spam-style headings—e.g. keywords for “thesis,” “write,” “where,” “topics,” “trusted,” “guide,” drug names, or porn tags.

    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
     tragedia
    -1.17
    produkt
    -1.09
    ****/
    -1.00
     sezonu
    -0.98
    举动
    -0.98
    wieder
    -0.97
    nicht
    -0.96
     animales
    -0.96
    pekte
    -0.96
     histó
    -0.95
    POSITIVE LOGITS
    illermo
    0.98
    最も
    0.94
    终于
    0.93
            
    0.92
                  
    0.89
    完整
    0.88
     für
    0.87
     konzent
    0.85
                    
    0.85
     unterstützt
    0.85
    Activations Density 0.004%

    No Known Activations