© 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. 5720
    Prev
    Next
    INDEX
    Explanations

    important considerations

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

    The neuron fires on the structural “breadcrumb” or metadata lines of forum posts—i.e. on the slashes, backslashes, quotes, and category/tag names that form the Discussion-in/posted-in/path 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
     two
    0.63
     =
    0.61
    ？
    0.61
    0.60
     +
    0.59
     with
    0.58
     only
    0.57
    ?
    0.56
    FromArgb
    0.55
     by
    0.54
    POSITIVE LOGITS
     secciones
    0.74
     Basics
    0.71
     basics
    0.70
    basics
    0.70
     بخش
    0.66
    fundamentals
    0.65
     sección
    0.65
    โครงการ
    0.65
    overview
    0.63
    sections
    0.63
    Activations Density 0.302%

    No Known Activations