© 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. 98396
    Prev
    Next
    INDEX
    Explanations

    UI fragments and controllers

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

    The neuron fires on programming‐style identifiers—especially UI component and layout class or file names (e.g. FrameLayout, Fragment, XML tags).

    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
     bluse
    -1.23
     lada
    -1.20
     idyl
    -1.20
     jurk
    -1.19
     hemd
    -1.16
    ремонт
    -1.13
     individuelles
    -1.13
     indisponible
    -1.13
     podían
    -1.10
    Наводи
    -1.10
    POSITIVE LOGITS
    }{{
    1.28
     while
    1.27
     with
    1.22
     without
    1.19
     so
    1.15
     after
    1.14
     showed
    1.04
     which
    1.01
     most
    1.00
    👋
    0.98
    Activations Density 0.076%

    No Known Activations