© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 16052
    Prev
    Next
    INDEX
    Explanations

    contrasting clauses start

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

    The neuron is highly selective for the curly apostrophe character (the ’ mark) in 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
     your
    0.55
     all
    0.54
     th
    0.53
     Mac
    0.51
    较
    0.50
     HH
    0.48
     IM
    0.48
     seren
    0.48
     l
    0.48
     QE
    0.48
    POSITIVE LOGITS
    Aside
    0.69
     Aside
    0.60
     ขณะ
    0.60
    Asimismo
    0.57
    Despite
    0.57
     Однако
    0.57
     رئيس
    0.56
     Несмотря
    0.56
    Opposition
    0.55
     بينما
    0.55
    Activations Density 0.140%

    No Known Activations