© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 103696
    Prev
    Next
    INDEX
    Explanations

    mutation and query

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

    programming terms related to mutations or data modification in code, particularly in GraphQL and database contexts.

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

    The neuron activates on code mentions of “mutation” (especially GraphQL mutation resolvers) and related resolver/query keywords.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ндарт
    -1.72
     fald
    -1.67
    ᱽ
    -1.66
    ︡
    -1.60
    cended
    -1.60
     awakeFromNib
    -1.55
     ESPEC
    -1.55
    “,
    -1.55
    鉺
    -1.54
    ♪♪
    -1.53
    POSITIVE LOGITS
     bijna
    1.58
     
    1.54
    -
    1.39
    ↵
    1.38
    &
    1.34
    </b>
    1.32
    renewcommand
    1.30
     tamamen
    1.30
     <
    1.28
     prawie
    1.27
    Activations Density 0.004%

    No Known Activations