© 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. 10-GEMMASCOPE-RES-131K
    7. 58699
    Prev
    Next
    INDEX
    Explanations

    geographical regions and scientific classifications

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

    website URL paths to different section categories like news, world, technology, politics, or education.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on URL path segments (and similar taxonomy/category labels) that denote site sections or topic categories (e.g. “world,” “asia,” “politics,” “education,” “technology,” taxonomic ranks, 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
     here
    -1.66
     are
    -1.63
     is
    -1.52
     
    -1.48
    c
    -1.46
    s
    -1.43
    r
    -1.40
    ly
    -1.40
     was
    -1.38
     the
    -1.36
    POSITIVE LOGITS
    Buna
    1.49
     Fußballspieler
    1.46
    شمار
    1.41
     but
    1.41
    忖
    1.40
    mainAxisSize
    1.39
    珅
    1.38
    Sự
    1.37
    ắ
    1.37
    řízen
    1.34
    Activations Density 0.374%

    No Known Activations