© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 14533
    Prev
    Next
    INDEX
    Explanations

    cleanliness and neatness

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

    The neuron reliably fires on mentions of cleaning or maintenance actions—especially on forms of “clean” (cleaning, cleaned) and related appliance/hood extractor contexts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    yaratan
    1.01
    nika
    0.96
    tensor
    0.89
     設計
    0.89
    payment
    0.88
    édération
    0.88
     Asamblea
    0.86
    ų
    0.86
    nál
    0.85
    設立
    0.84
    POSITIVE LOGITS
     cleaning
    2.06
     cleans
    2.05
     cleansing
    2.03
     cleanliness
    1.76
     cleanser
    1.75
     Cleaning
    1.73
     cleaners
    1.73
     disinfecting
    1.72
    🧼
    1.69
     nettoyage
    1.68
    Activations Density 2.187%

    No Known Activations