© 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. 22-GEMMASCOPE-RES-131K
    4. 68155
    Prev
    Next
    INDEX
    Explanations

    common prefixes of phrases

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

    This neuron detects capitalized or proper‐noun tokens—i.e. words or abbreviations appearing with initial uppercase letters (often sentence-initial words, names, or acronyms).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     otros
    -1.29
    <bos>
    -1.27
    暧
    -1.25
     aston
    -1.23
    そこで
    -1.21
    この
    -1.17
     Arrondissement
    -1.16
    長さ
    -1.15
     pittores
    -1.14
    我們
    -1.13
    POSITIVE LOGITS
    -
    1.39
     always
    1.34
    為に
    1.26
     muszą
    1.23
     wasn
    1.23
     people
    1.23
     something
    1.22
    ,
    1.20
     dużo
    1.20
    izing
    1.20
    Activations Density 0.094%

    No Known Activations