© 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. 118752
    Prev
    Next
    INDEX
    Explanations

    Protocol Buffers or service configurations

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

    the word "Buffer" or "Buffers" in technical contexts, particularly in Protocol Buffers references.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron is spotting word endings—that is, common suffix morphemes (like ‑ers, ‑ed, ‑ing, ‑ions, etc.).

    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
     billeder
    -1.47
     not
    -1.41
     bahkan
    -1.27
     deriva
    -1.23
     related
    -1.23
     plan
    -1.23
     for
    -1.23
    lepaskan
    -1.23
     your
    -1.22
     new
    -1.22
    POSITIVE LOGITS
    Another
    1.61
     facendo
    1.53
     takže
    1.42
    если
    1.39
    リエーション
    1.38
     informací
    1.36
    อีก
    1.32
    сутствие
    1.30
    最后一个
    1.27
     उन्ह
    1.27
    Activations Density 0.102%

    No Known Activations