© 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. 22-GEMMASCOPE-RES-131K
    7. 68772
    Prev
    Next
    INDEX
    Explanations

    web application features

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

    The neuron activates strongly on PWA-related manifest and service‐worker terms (e.g. keys like “service_worker,” “start_url,” “display,” “icons,” “offline,” 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_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
    lasyon
    -1.17
    iveau
    -1.14
     Vorstand
    -1.12
     smaller
    -1.11
     Př
    -1.07
    ='';
    -1.06
    另一
    -1.06
    relenting
    -1.05
     $\
    -1.04
    录取
    -1.03
    POSITIVE LOGITS
    頰
    1.28
     sayesinde
    1.16
     però
    1.13
    一来
    1.13
    ૉ
    1.10
    Literatuur
    1.09
     世代
    1.08
     checkup
    1.07
    Aunque
    1.06
     ɛ
    1.05
    Activations Density 0.004%

    No Known Activations