© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 69562
    Prev
    Next
    INDEX
    Explanations

    political entities

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

    This neuron detects uppercase labels or datelines (e.g. city names or “Share”) immediately followed by a colon and a numeric value.

    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
     ehemalige
    -1.79
     uſed
    -1.79
     Vielfalt
    -1.69
     vorhand
    -1.65
     къде
    -1.61
     Gläub
    -1.59
    tranh
    -1.58
     ähnliche
    -1.56
    In
    -1.52
     natureza
    -1.50
    POSITIVE LOGITS
    \
    1.76
    </i>
    1.61
    owning
    1.52
    はじ
    1.50
    ־ה
    1.44
     fête
    1.43
     deciso
    1.41
     wir
    1.40
     cedera
    1.40
     if
    1.39
    Activations Density 0.018%

    No Known Activations