© 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. 11470
    Prev
    Next
    INDEX
    Explanations

    technical specifications and definitions

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

    This neuron responds to formal patent‐style headings and boilerplate phrases (e.g. “International Search Report,” “the present invention,” “According to…,” “なお”) that mark sections or standard legal/disclosure language.

    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
     मर्दों
    0.98
    ମ୍
    0.94
     Hispan
    0.88
    绾
    0.88
    ାର
    0.87
    思います
    0.86
     sexes
    0.85
    ើស
    0.83
    ំ
    0.83
     हार्म
    0.83
    POSITIVE LOGITS
    to
    1.00
    nt
    0.91
    treat
    0.90
    tor
    0.89
    J
    0.89
    ta
    0.88
    -
    0.86
    it
    0.84
    ाना
    0.84
    gu
    0.82
    Activations Density 0.007%

    No Known Activations