© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26
    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 114046
    Prev
    Next
    INDEX
    Explanations

    programming libraries and tools

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

    the term "appcompat" or related Android support library references in code.

    oai_token-act-pair · claude-4-5-sonnet
    Triggered by @bfgcn7frf5

    The neuron fires on mentions of external APIs, libraries, and system‐service identifiers (e.g. “androidx.appcompat,” “OpenSSL,” or “getSystemService”).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    ",
    -2.78
    ',
    -2.38
    {
    -2.36
    ’
    -2.30
    o
    -2.20
    。“
    -2.17
    l
    -2.11
    ”—
    -2.06
    眉毛
    -2.06
    ጵ
    -1.99
    POSITIVE LOGITS
    寕
    3.00
    睒
    2.80
    
    2.58
    甃
    2.33
    which
    2.33
     marqués
    2.31
    ಞ
    2.30
    ꯹
    2.30
    鹮
    2.28
    ᜀ
    2.28
    Activations Density 0.005%

    No Known Activations