© 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. 21137
    Prev
    Next
    INDEX
    Explanations

    proper names and foreign text

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

    This neuron detects numeric values—especially numbers in parentheses such as years, reference numbers, or measurements.

    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
    </h2>
    -1.84
     it
    -1.73
    </b>
    -1.67
     azonban
    -1.53
     Thus
    -1.52
     such
    -1.52
     said
    -1.47
     semblent
    -1.47
     Therefore
    -1.45
    які
    -1.43
    POSITIVE LOGITS
     другую
    1.63
    unno
    1.48
     oud
    1.40
     levé
    1.40
    modium
    1.37
    odine
    1.34
    墙壁
    1.34
     وه
    1.34
    儸
    1.32
    鰨
    1.30
    Activations Density 0.260%

    No Known Activations