© 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-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 5467
    Prev
    Next
    INDEX
    Explanations

    broken down into categories

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

    This neuron fires on sentences describing lodging or real-estate amenities—specifically on verbs like “offers,” “features,” or “enjoys” when listing property or hotel services.

    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
     automatiquement
    0.84
    렬
    0.82
    rosine
    0.78
     ouvrir
    0.77
    groovy
    0.76
     améli
    0.75
     impulso
    0.75
     bleue
    0.73
    roboto
    0.73
    >`
    0.71
    POSITIVE LOGITS
    یکا
    0.93
     مَا
    0.86
    させていただきます
    0.82
    ўся
    0.80
    させて頂きます
    0.79
    тивность
    0.78
    اويه
    0.76
    ীত
    0.76
    闊
    0.75
     экспер
    0.73
    Activations Density 0.002%

    No Known Activations