© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 7245
    Prev
    Next
    INDEX
    Explanations

    std::cout << endl

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

    This neuron detects programming‐style punctuation and operators (e.g. stream insertion “<<”, string quotes “= "…", concatenation “+=”, “echo”, etc.) 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
    ['
    0.81
    {'
    0.73
    +="
    0.73
    ["
    0.73
    =="
    0.72
    Со
    0.70
    +'
    0.69
    ==="
    0.69
    =='
    0.66
    +"
    0.65
    POSITIVE LOGITS
     (!_
    0.80
     "/"
    0.74
     SKAction
    0.73
     "("
    0.71
     tendrás
    0.70
     ":
    0.69
     "|"
    0.69
     također
    0.69
     scopo
    0.68
     endl
    0.68
    Activations Density 0.003%

    No Known Activations