© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 8177
    Prev
    Next
    INDEX
    Explanations

    sports leagues expansion and officials

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

    The neuron fires on uppercase abbreviations and acronyms (often with periods), especially names of leagues, organizations, or similar all-caps initialisms.

    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
    我们
    -1.47
    有不少
    -1.45
     chrétienne
    -1.28
     orien
    -1.25
    Ārējās
    -1.21
     cults
    -1.20
     وصلة
    -1.19
     we
    -1.19
    PDIR
    -1.19
     udało
    -1.17
    POSITIVE LOGITS
     its
    1.76
     itself
    1.52
    を行い
    1.40
     countries
    1.30
     dejen
    1.29
     would
    1.29
     démocr
    1.29
    :“……”
    1.25
     написать
    1.24
    ventilador
    1.24
    Activations Density 0.204%

    No Known Activations