© 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. 9372
    Prev
    Next
    INDEX
    Explanations

    django

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

    The neuron strongly activates on tokens containing non-English or accented Unicode characters (e.g. high-ASCII/diacritic sequences).

    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
    เหตุ
    0.57
    OrElse
    0.56
     ▸
    0.55
    ções
    0.53
     ›
    0.52
     달리
    0.52
     ocorr
    0.52
     difer
    0.51
     Toler
    0.51
    で
    0.50
    POSITIVE LOGITS
    specification
    0.61
    ython
    0.60
    字符串
    0.59
     UNIVERSITY
    0.59
    rupt
    0.58
     raced
    0.58
    اث
    0.57
     करेक्ट
    0.57
     እየሱስ
    0.57
    ycles
    0.57
    Activations Density 0.051%

    No Known Activations