© 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. 28227
    Prev
    Next
    INDEX
    Explanations

    UI backdrop and cancel options

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

    The neuron detects terms related to closing or dismissing a modal dialog (e.g. backdrop, mask, clicking outside, pressing escape, static).

    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
     precisam
    -1.21
    Niveau
    -1.14
     sabem
    -1.13
     necessità
    -1.13
     provavelmente
    -1.12
     encontram
    -1.11
     terão
    -1.06
     Harz
    -1.05
     servem
    -1.05
     singur
    -1.04
    POSITIVE LOGITS
    archiviato
    1.05
     joven
    1.02
    冶
    1.01
    optarg
    1.00
    IENCE
    1.00
     envi
    0.98
    GroupName
    0.98
    沼
    0.98
    ProgressDialog
    0.97
    ască
    0.97
    Activations Density 0.001%

    No Known Activations