© 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. 130577
    Prev
    Next
    INDEX
    Explanations

    events and numbered acts

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

    The neuron detects numeric references—especially years and statute or act numbers in legal citations.

    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
     how
    -2.28
     this
    -2.23
     different
    -2.13
     various
    -2.03
     two
    -1.86
     with
    -1.83
     as
    -1.82
     a
    -1.82
     such
    -1.82
     these
    -1.81
    POSITIVE LOGITS
     România
    1.64
     représenter
    1.57
    罍
    1.54
    chten
    1.52
    ཌ
    1.52
    iptables
    1.48
    siert
    1.47
     😉
    1.47
    tió
    1.46
     diminuer
    1.45
    Activations Density 0.027%

    No Known Activations