© 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. 31-GEMMASCOPE-2-RES-16K
    7. 8157
    Prev
    Next
    INDEX
    Explanations

    style properties

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

    The neuron fires on the colon punctuation used to introduce CSS pseudo‐classes or similar code delimiters.

    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
     =
    0.61
    igel
    0.60
    ="
    0.55
     abre
    0.55
     cellspacing
    0.54
    evan
    0.54
     Tall
    0.54
     squeeze
    0.53
    ację
    0.53
     =(
    0.53
    POSITIVE LOGITS
     કર્મ
    0.56
    ExternalTaskPojo
    0.52
     বিলম্ব
    0.52
    มั่น
    0.52
    竝
    0.52
     وبركاته
    0.51
    Bers
    0.51
    爨
    0.51
     delineation
    0.51
    និយ
    0.50
    Activations Density 0.012%

    No Known Activations