© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 4298
    Prev
    Next
    INDEX
    Explanations

    legal terms like agreement, license, contract

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

    This neuron detects formal legal-defined terms or section headings (e.g., “Claim,” “Site,” “Terms,” “Agreement,” “Trademarks”) in contract‐style text.

    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
     poprzez
    0.71
    视为
    0.71
    甚至是
    0.70
     hatta
    0.70
    更加
    0.68
     melalui
    0.68
    更好地
    0.68
     meningkatkan
    0.67
     davantage
    0.67
     through
    0.64
    POSITIVE LOGITS
     เริ่ม
    0.61
    ርስዎ
    0.58
     ขาย
    0.57
    figuration
    0.55
     বৎসর
    0.54
     งวด
    0.54
     xảy
    0.52
     นั่น
    0.52
     વરસ
    0.52
     약
    0.52
    Activations Density 0.001%

    No Known Activations