© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 12654
    Prev
    Next
    INDEX
    Explanations

    code before punctuation

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

    This neuron fires on programming‐language syntax elements—e.g. parentheses, operators (==, =>), comparison tokens, method‐call punctuation, and other code‐specific symbols.

    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
    {}
    0.56
     {}
    0.54
     [
    0.52
     [%
    0.49
     integer
    0.49
     penge
    0.48
     பெரி
    0.48
    డిన
    0.47
     অবস্থানের
    0.47
    وابة
    0.47
    POSITIVE LOGITS
    ($"
    1.26
     $"{
    1.09
     $"
    1.06
    ($"{
    1.06
     JsonConvert
    0.89
     $@
    0.88
    nameof
    0.83
     Mathf
    0.82
     Convert
    0.79
    GUILayout
    0.75
    Activations Density 0.005%

    No Known Activations