© 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. 3
    Prev
    Next
    INDEX
    Explanations

    parentheses and list items

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

    The neuron flags punctuation used to set off parenthetical asides or clause breaks (e.g. em-dashes, ellipses, parentheses).

    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
     Kemudian
    0.24
     อ่า
    0.20
    ↵↵
    0.20
    [
    0.20
    <0x0D>
    0.20
     ン
    0.19
    </
    0.19
    0.19
    Kemudian
    0.19
    0.19
    POSITIVE LOGITS
    including
    0.32
    especially
    0.31
    doesn
    0.30
    literally
    0.28
    whether
    0.28
    particularly
    0.27
    at
    0.26
    albeit
    0.26
    sometimes
    0.26
    often
    0.26
    Activations Density 0.722%

    No Known Activations