© 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. 73791
    Prev
    Next
    INDEX
    Explanations

    levels and rooms

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

    The neuron fires on real-estate listing jargon—particularly structural and room-type terms (e.g. main, floor, room, bedroom, bathroom, garage) when describing property layouts.

    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
     to
    -1.45
     each
    -1.14
     it
    -1.10
     their
    -0.97
     since
    -0.94
     from
    -0.93
     its
    -0.92
     AppColors
    -0.89
    otá
    -0.88
    especially
    -0.88
    POSITIVE LOGITS
     ejes
    1.18
     Públicas
    1.02
     औ
    0.98
    Теле
    0.98
    фраструк
    0.98
    jący
    0.97
    uose
    0.96
    ГЛА
    0.95
     dynamite
    0.95
    Рус
    0.94
    Activations Density 0.009%

    No Known Activations