Files
LLM-from-scratch/README.md
T
2025-05-22 18:13:01 +08:00

49 lines
2.0 KiB
Markdown
Executable File
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
# LLM-Dolly
LLM-Dolly is a custom LLM built from scratch
持续优化更新中...
## 🗓️ 更新日志
#### 2025-05-22
- 📝 添加支持MoE模型,训练资源占用不稳定,测试模型Experts=8, experts_per_tok=4,GPU显存从60%-94%跳动,单个GPU利用率0%-100%跳动。当调小batch_size后,显存占用稳定在90%,单个GPU利用率稳定在90%以上,偶现30%的利用率。
#### 2025-05-20
- 📝 添加支持jsonl文件训练.
#### 2025-05-19
- 📝 添加支持使用deepspeed训练代码,测试训练中最大batch_size提升38%,训练速度提升9.6%.
#### 2025-05-16
- 📝 添加dolly_llm的预训练代码,进行一次预训练测试:模型0.6B,语料500M,46GB*4显卡.
#### 2025-05-14
- ✅ 将dolly_llm作为pip包,进行安装
#### 2025-05-09
- 📝 优化[RMSNorm](https://arxiv.org/pdf/1910.07467)、MLP、[RoPE](https://arxiv.org/pdf/2104.09864)代码。
#### 2025-05-07
- 📝 使用transformers格式规范modeling和configuration,并设计修改v0.1版modeling_dolly,参数量11.5B。
#### 2025-05-06
- 📝 实现 configuration_dolly类,以及添加v0.0版modeling_dolly。
#### 2025-04-30
- 📝 添加Tokenizer的[BBPE](https://arxiv.org/pdf/1909.03341)方式训练。
#### 2025-04-29
- 📝 添加tokenizer构建代码,支持sentencepiece和transfomers的[BPE](https://arxiv.org/pdf/1508.07909),支持从文本构建和从已有的tokenzier构建。
#### 2025-04-24
- ✅ 测试从transformers构建自定义的LLM模型结构。
## 致谢
特别感谢以下资源和文章的帮助:
- [Huggingface transformers](https://github.com/huggingface/transformers)
- [Qwen](https://huggingface.co/Qwen)
- [Neural Machine Translation of Rare Words with Subword Units](https://arxiv.org/pdf/1508.07909)
- [Neural Machine Translation with Byte-Level Subwords](https://arxiv.org/pdf/1909.03341)
- [Root Mean Square Layer Normalization](https://arxiv.org/pdf/1910.07467)
- [RoPE RoFormer: Enhanced Transformer with Rotary Position Embedding](https://arxiv.org/pdf/2104.09864)