© 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. 109283
    Prev
    Next
    INDEX
    Explanations

    virus spread and mutation types

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

    descriptive text about viruses, diseases, and pathogens.

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

    This neuron detects technical descriptions of infectious diseases, especially formal definitions and attributes of a virus or disease (e.g. “X is a … disease caused by a virus and is spread by…”).

    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
    siune
    -0.97
    糖尿病
    -0.90
    出版
    -0.89
     hoppas
    -0.88
    ブラ
    -0.88
     Funktion
    -0.87
     blessés
    -0.85
    Leipzig
    -0.85
    چرا
    -0.85
    قیمت
    -0.85
    POSITIVE LOGITS
     spread
    1.80
     Spread
    1.47
     attacking
    1.41
     spreads
    1.37
     attack
    1.34
     wre
    1.20
     affecting
    1.19
    Attacking
    1.19
    attack
    1.17
    spread
    1.16
    Activations Density 0.083%

    No Known Activations