© 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. 1600
    Prev
    Next
    INDEX
    Explanations

    class declarations

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

    This neuron specifically picks out standalone numeric or punctuation‐only code tokens (e.g. numbers, parentheses, and periods) in code snippets.

    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
    nullptr
    0.71
     nullptr
    0.70
     风
    0.68
     invigorating
    0.64
     वैशिष्ट
    0.63
    ẏ
    0.61
     বজ্র
    0.61
     underpinned
    0.61
    奻
    0.61
     मजेदार
    0.60
    POSITIVE LOGITS
     boolean
    1.59
    boolean
    1.58
     int
    1.52
    int
    1.51
     String
    1.50
    String
    1.48
     StringBuilder
    1.42
    ArrayList
    1.35
     ArrayList
    1.34
    StringBuffer
    1.28
    Activations Density 0.012%

    No Known Activations