© 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. 84116
    Prev
    Next
    INDEX
    Explanations

    closing eyes, gaps, doors, books

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

    The main thing this neuron does is detect references to closing or shutting one’s eyes.

    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
     Selecciona
    -1.13
    ↵↵
    -1.10
     deras
    -1.04
     轉
    -1.03
    眼光
    -1.02
     sözler
    -1.00
    也想
    -1.00
    Estructura
    -1.00
    corro
    -0.98
    fmf
    -0.98
    POSITIVE LOGITS
     closes
    1.48
     closed
    1.34
    ção
    1.29
     shut
    1.20
    fclose
    1.19
    -</
    1.15
    shut
    1.13
    esperienza
    1.13
     cierra
    1.10
     doors
    1.09
    Activations Density 0.038%

    No Known Activations