© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8670
    Prev
    Next
    INDEX
    Explanations

    quotation marks and brackets

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

    This neuron detects punctuation marks—especially sentence boundary symbols such as periods, commas, colons, and question marks.

    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_31_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
    
    0.57
     diversos
    0.54
     sham
    0.54
     various
    0.53
     unen
    0.51
    \/
    0.51
     burgeoning
    0.50
    ífer
    0.50
    ULAR
    0.50
     ብዙውን
    0.50
    POSITIVE LOGITS
    ：“
    2.44
     "...
    2.41
     “…
    2.38
    :“
    2.31
     “[
    2.20
     “
    2.19
    :"
    2.17
    ：「
    2.17
     "[
    2.16
     “‘
    2.16
    Activations Density 0.405%

    No Known Activations