© 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. 75683
    Prev
    Next
    INDEX
    Explanations

    analysis of pollutants and elements

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

    chemical compounds and scientific notation in technical/environmental documents.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron activates on technical chemical contaminant acronyms or formula-like abbreviations (e.g. PCDD/PCDF, PCB, etc.).

    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
    blauch
    -0.82
     válvulas
    -0.79
    urally
    -0.77
    医学
    -0.77
    僕
    -0.73
     puppets
    -0.71
     sizin
    -0.70
     gerar
    -0.69
     vindo
    -0.69
    AGON
    -0.69
    POSITIVE LOGITS
     analysis
    1.28
    analysis
    1.15
     analyses
    1.13
    分析
    1.07
     विश्लेषण
    1.06
     Analysis
    1.05
     metals
    1.03
    Analysis
    1.03
    analyze
    1.01
     analy
    0.98
    Activations Density 0.021%

    No Known Activations