Show HN: RLHF、注意力与 LLM 评估交互笔记

原标题:Show HN: Interactive notes on RLHF, attention, LoRA and LLM evals
42AI 研判分

开发者发布了一系列交互式技术笔记,涵盖 RLHF、注意力机制、LoRA 及 LLM 评估等核心概念。该资源通过允许用户修改输入来直观理解工程原理,适合 AI 从业者作为辅助学习工具。

报道正文

本文目录3 个章节

Engineering ideas you understand by changing the inputs.

Evals

The map of evalsStart here

  1. Is 82% better than 80%?
  2. Exact match and F1
  3. Precision, recall and reciprocal rank
  4. nDCG@k
  5. pass@k
  6. LLM as judge
  7. Pairwise preference and Elo
  8. Human agreement and kappa
  9. RAG evals
  10. Agent evals
  11. Online evals and A/B tests
  12. Contamination and Goodhart

Machine learning

Search and ranking

Notes I keep for myself, added as I learn. Found a mistake? · Source · Analytics choice

Allow Google Analytics to count page views? It sets cookies. Nothing else on this site tracks you.