© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 1800
    Prev
    Next
    INDEX
    Explanations

    The words "include" or "what to include"

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

    This neuron detects document structure markers—most notably list or section numbers and their accompanying punctuation (e.g. “1/”, “2/”, colons, commas, and periods) used in outlines, headings, and checklists.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    имость
    -0.86
     correspondance
    -0.84
    "-"
    -0.83
    攞
    -0.82
     ดำ
    -0.81
    諷
    -0.81
    しか
    -0.80
    walde
    -0.80
    莅
    -0.79
    istern
    -0.78
    POSITIVE LOGITS
     include
    2.28
    Include
    1.89
     Include
    1.88
     included
    1.84
    include
    1.84
     inclusion
    1.72
     вклю
    1.60
     your
    1.57
     incluir
    1.53
     включа
    1.46
    Activations Density 0.036%

    No Known Activations