© 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. 10076
    Prev
    Next
    INDEX
    Explanations

    adjectives before nouns

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

    The neuron is tuned to pick out e-commerce listing keywords—brand names, product categories, quantities, and specs that appear in item titles or headers.

    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
    novation
    0.72
    āo
    0.71
    With
    0.68
     With
    0.67
    还有
    0.66
     chứ
    0.58
    rospection
    0.57
     所
    0.57
    nahi
    0.57
    Id
    0.57
    POSITIVE LOGITS
    ステンレス
    0.72
     lightweight
    0.65
     aluminium
    0.63
     Lightweight
    0.62
     alumin
    0.61
     زيت
    0.61
    小型
    0.61
     barato
    0.61
     motocicleta
    0.61
     earthenware
    0.61
    Activations Density 0.162%

    No Known Activations