Ruida Docs
search
⌘Ctrlk
GitBook Assistant
GitBook Assistant
Working...Thinking...
GitBook Assistant
Good morning

I'm here to help you with the docs.

⌘Ctrli
AI Based on your contextquestion-circle
Ruida Docs
  • a. 基础知识
  • b. PyTorch
  • c. LLM 基础
  • d. 分布式训练
  • e. Pre-Training
  • f. Post-Training
  • g. LLM Inference
  • h. Agent
  • i. 主流大模型技术
    • 1. GPT
    • 10. Kimi
    • 11. GLM
    • 12. Nemotron
    • 2. Llama
    • 3. Gemini
    • 4. DeepSeek
    • 5. Qwen
    • 6. OLMo
    • 7. Gemma
    • 8. Mistral
    • 9. SmolLM
    • LLM Frontier
    • LLM Reasoning
  • j. 其他
gitbookPowered by GitBook
block-quoteOn this pagechevron-down
  1. i. 主流大模型技术

2. Llama

介绍 Llama 系列模型的架构与训练方法

  • Llama 主要技术:GQA、RMSNorm、SwiGLU FFN、RoPE

  • 参考资料

    • 帖子

      • Llama 2 and FlashAttention 2 - by Sebastian Raschka, PhDarrow-up-right

      • Llama 4: The Challenges of Creating a Frontier-Level LLMarrow-up-right

    • 论文

      • Llama:LLaMA: Open and Efficient Foundation Language Modelsarrow-up-right

      • Llama 2:Llama 2: Open Foundation and Fine-Tuned Chat Modelsarrow-up-right

      • Llama 3:The Llama 3 Herd of Modelsarrow-up-right

      • Llama 3.1:https://ai.meta.com/blog/meta-llama-3-1/

      • Llama 4:The Llama 4 herd: The beginning of a new era of natively multimodal AI innovationarrow-up-right

hashtag
Llama

hashtag
Llama 2

hashtag
Llama 3

hashtag
Llama 4

Previous12. Nemotronchevron-leftNext3. Geminichevron-right

Last updated 46 minutes ago

Was this helpful?

Created By Ruida

Was this helpful?