N-Gram House

Tag: learned embeddings

Positional Encoding in Transformers: Sinusoidal vs Learned for LLMs

Positional Encoding in Transformers: Sinusoidal vs Learned for LLMs

Sinusoidal and learned positional encodings were early solutions for transformers, but modern LLMs now use RoPE and ALiBi for better long-context performance. Learn why and how these techniques evolved.

Categories

  • History (50)
  • Machine Learning (14)

Recent Posts

Measuring Hallucination Rate in Production LLM Systems: Key Metrics and Real-World Dashboards Jan, 5 2026
Measuring Hallucination Rate in Production LLM Systems: Key Metrics and Real-World Dashboards
Token Probability Calibration in Large Language Models: How to Make AI Confidence More Reliable Aug, 10 2025
Token Probability Calibration in Large Language Models: How to Make AI Confidence More Reliable
Hybrid Search for RAG: Boost LLM Accuracy with Semantic and Keyword Retrieval Dec, 7 2025
Hybrid Search for RAG: Boost LLM Accuracy with Semantic and Keyword Retrieval
Guardrail-Aware Fine-Tuning to Reduce Hallucination in Large Language Models Feb, 1 2026
Guardrail-Aware Fine-Tuning to Reduce Hallucination in Large Language Models
Choosing Opinionated AI Frameworks: Why Constraints Boost Results Jan, 20 2026
Choosing Opinionated AI Frameworks: Why Constraints Boost Results

Menu

  • About
  • Terms of Service
  • Privacy Policy
  • CCPA
  • Contact

© 2026. All rights reserved.