© 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. 11433
    Prev
    Next
    INDEX
    Explanations

    special characters and non-English letters

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

    The neuron is tuned to detect numeric tokens and numeric expressions (e.g. dates, counts, percentages) 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
    //{
    0.54
    ++){
    0.51
    {[
    0.48
     BELOW
    0.47
    鸴
    0.47
    рто
    0.46
    试
    0.46
     sebelah
    0.46
     pripad
    0.46
    %(
    0.46
    POSITIVE LOGITS
     Suisse
    0.62
     Lint
    0.59
    addon
    0.55
     Cumbria
    0.55
     adm
    0.54
     Bristol
    0.54
     こ
    0.53
     Midlands
    0.52
     फैसला
    0.51
    ിലാ
    0.51
    Activations Density 0.001%

    No Known Activations