© 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. 10519
    Prev
    Next
    INDEX
    Explanations

    code block delimiters

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

    The neuron is tuned to pick up code‐style tokens and numeric or indexing expressions (e.g. lindex, argv, expr/rand, decimal values) rather than natural language words.

    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
     concerning
    0.57
    rapati
    0.56
    YN
    0.54
    erk
    0.52
     Tina
    0.51
    걱
    0.51
    arev
    0.50
     ")[
    0.49
    reat
    0.49
    هار
    0.49
    POSITIVE LOGITS
     {(
    1.01
     {{
    0.96
     {-
    0.96
     {'
    0.87
    {(
    0.86
    {\|
    0.85
    {-
    0.82
    {-\
    0.82
     {[
    0.80
    {$\
    0.80
    Activations Density 0.006%

    No Known Activations