© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 30027
    Prev
    Next
    INDEX
    Explanations

    United States Court of Appeals

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

    The neuron fires on tokens that look like technical identifiers or formal headings—i.e. words in all‐caps or with underscores/CamelCase (e.g. method names like “has_content,” license “GNU,” or court headings like “UNITED STATES COURT OF APPEALS”).

    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_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    olph
    -0.78
     Kiri
    -0.76
    ésor
    -0.72
    🤏
    -0.70
    orcid
    -0.69
    Kiri
    -0.69
    bitur
    -0.69
    GREY
    -0.68
    lidean
    -0.68
    ückt
    -0.68
    POSITIVE LOGITS
     isEqual
    0.69
     Hef
    0.65
    Acknowledgment
    0.63
     Сло
    0.62
     Continue
    0.62
     תה
    0.61
    chale
    0.61
    нию
    0.60
    hyphen
    0.60
    名が
    0.60
    Activations Density 0.063%

    No Known Activations