© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 2603
    Prev
    Next
    INDEX
    Explanations

    object properties

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

    The neuron fires on tokens that come from side-bars or metadata snippets—things like parenthetical annotations, uppercase acronyms, named entities, list labels, and other “aside” markers rather than ordinary running 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
     Algebras
    0.60
    Seqlist
    0.55
     LOCCTR
    0.55
     প্রবৃত্ত
    0.55
    GBuf
    0.53
    ๚
    0.51
    ناط
    0.51
     BasicContainer
    0.51
     vielfält
    0.50
    หลากหลาย
    0.50
    POSITIVE LOGITS
    (
    0.79
     (
    0.71
     see
    0.59
     you
    0.56
    ($
    0.56
     get
    0.53
     [
    0.53
     don
    0.51
    1
    0.50
    2
    0.50
    Activations Density 0.029%

    No Known Activations