© 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. 22-GEMMASCOPE-RES-131K
    4. 10889
    Prev
    Next
    INDEX
    Explanations

    subtype classification

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

    The neuron responds strongly to specialized genetic/virology terminology—especially words naming subtypes, lineages, recombination events, and other technical biological classification terms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     chré
    -1.90
    皈
    -1.54
     atque
    -1.42
    érios
    -1.40
     يعمل
    -1.32
     パソコン
    -1.30
     religieuses
    -1.30
     empfiehlt
    -1.30
     растение
    -1.30
    =\"
    -1.30
    POSITIVE LOGITS
    ֶּ
    1.85
    ִּ
    1.58
    ּוֹ
    1.55
    ֵּ
    1.51
    ְר
    1.48
     مباراة
    1.43
    ְּ
    1.42
     сообщили
    1.41
    Niet
    1.39
    hoka
    1.36
    Activations Density 0.258%

    No Known Activations