© 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. 10923
    Prev
    Next
    INDEX
    Explanations

    their or they followed by common phrases

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

    The neuron is strongly triggered by numeric measurements—especially floating‐point numbers and other quantity values in the 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_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
     %.
    0.52
     Practice
    0.51
     &'
    0.51
     practice
    0.50
     amp
    0.49
    StringVar
    0.48
     %
    0.48
    practice
    0.48
     Languages
    0.48
    %.
    0.47
    POSITIVE LOGITS
     অবশ্যই
    0.64
     deemed
    0.62
     conceivable
    0.61
    FEB
    0.60
    icent
    0.59
    0.55
     implies
    0.55
     swayed
    0.55
     बिल्कुल
    0.54
     amounting
    0.54
    Activations Density 0.003%

    No Known Activations