© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128611
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on standalone programming-related identifiers—like library, framework, or file-format names (e.g. “Json,” “Xml,” “React,” “Razor,” “proto,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    XML, JSON, React, proto serialization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     like
    -1.38
    “
    -1.33
     more
    -1.31
     its
    -1.20
     for
    -1.19
    ‘
    -1.14
     others
    -1.13
    <b>
    -1.10
     from
    -1.09
     muna
    -1.09
    POSITIVE LOGITS
    pantalones
    1.49
    Kemudian
    1.31
    zapatos
    1.30
    ஒ
    1.30
    Çok
    1.27
    Tils
    1.27
    بله
    1.25
     goû
    1.25
    pergillus
    1.23
     encargado
    1.23
    Activations Density 0.001%

    No Known Activations