© 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. 7107
    Prev
    Next
    INDEX
    Explanations

    Django objects

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

    The neuron strongly activates on code‐snippet tokens—programming identifiers, filenames (e.g. *.py), library names (e.g. Django), and the numeric literals often seen in code.

    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
    ícios
    0.64
    鹛
    0.57
    ViewSet
    0.56
    wsgi
    0.55
    WS
    0.55
    urlpatterns
    0.55
    یکا
    0.54
     পশ্চিমবঙ্গ
    0.53
    ws
    0.52
    API
    0.51
    POSITIVE LOGITS
     วัด
    0.55
    👕
    0.54
    histogram
    0.53
     cố
    0.52
     Kiel
    0.51
     softens
    0.51
    👔
    0.50
     Dus
    0.49
     Quir
    0.49
    गिल
    0.48
    Activations Density 0.001%

    No Known Activations