© 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. 5983
    Prev
    Next
    INDEX
    Explanations

    specific

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

    This neuron detects the openings of square‐bracketed parenthetical or asides (i.e. the “\[” token and the content immediately following it).

    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
     ちょっと
    0.70
     นิด
    0.68
     ไม่มี
    0.67
     VALUE
    0.65
    >";
    0.64
     +
    0.63
     こんな
    0.62
     ชีวิต
    0.61
     WILL
    0.61
     +,
    0.60
    POSITIVE LOGITS
    ,]
    0.66
    sic
    0.64
    laughs
    0.63
    الية
    0.60
    …。
    0.60
    Laugh
    0.59
    фіка
    0.58
    。[
    0.57
    .]
    0.57
    :]
    0.57
    Activations Density 0.012%

    No Known Activations