Embeddings

What is Agency? A View from Autonomy Theory

Back
1
Total Embedding Vectors
1
Documents Embedded
1
Period-Aware
1
Methods Used
Document Embeddings 1 documents
Pipeline
period_aware sentence-transformers/all-MiniLM-L6-v2
384
Dimensions
1
Chunks
Modern 1950 2000 Modern
Default model - insufficient metadata for period detection
50% selection confidence
About Embeddings
What are Embeddings?

Embeddings are dense vector representations of text that capture semantic meaning. Similar texts have similar embeddings, enabling semantic search and comparison.

Period-Aware Models
  • Pre-1850: all-mpnet-base-v2 (768 dims)
  • 1850-1950: all-mpnet-base-v2 (768 dims)
  • 1950-2000: all-MiniLM-L6-v2 (384 dims)
  • 2000+: all-roberta-large-v1 (1024 dims)
Usage

Period-aware embeddings select models optimized for historical eras, enabling accurate semantic comparison across time periods.