© 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. 22-GEMMASCOPE-RES-131K
    7. 71430
    Prev
    Next
    INDEX
    Explanations

    possessive markers like 's

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

    The neuron fires on possessive markers—e.g. English ’s or the Chinese genitive particle “的.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     one
    -1.13
    cije
    -1.08
     another
    -1.04
     inkluder
    -0.88
     itself
    -0.85
    的那
    -0.85
     Wars
    -0.84
    噻
    -0.84
     最
    -0.83
    ريع
    -0.82
    POSITIVE LOGITS
     leurs
    1.09
    affaire
    1.01
    による
    1.01
    굳
    1.00
     cuyas
    0.93
    うちの
    0.93
    resiasi
    0.91
    들의
    0.91
    がありますが
    0.90
     część
    0.90
    Activations Density 0.064%

    No Known Activations