© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 86668
    Prev
    Next
    INDEX
    Explanations

    sentence start after punctuation

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

    This neuron fires on tokens that mark major segment or sentence boundaries—i.e. the words or punctuation that signal the start or end of a new sentence or section.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     when
    -1.60
    ])):
    -1.38
     only
    -1.27
     before
    -1.25
     сні
    -1.24
     just
    -1.23
    ])){
    -1.18
    </h2>
    -1.16
    ">${
    -1.16
    just
    -1.14
    POSITIVE LOGITS
     всё
    1.34
     wszystkie
    1.23
     эмоциона
    1.21
     الذي
    1.20
     wszystkich
    1.16
    pastel
    1.16
     nyní
    1.15
     этого
    1.12
     měl
    1.11
     calefacción
    1.11
    Activations Density 0.109%

    No Known Activations