Repository Issue Activity (beta)

#1017 Improvement to SimpleTokenizer

#435 RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! in ch05/01_main-chapter-code/gpt_generate.py

#1061 Bug in Ch03 - Bonus02 - PyTorch's scaled dot product attention without FlashAttention

#1052 Add type hints to core modules for improved code quality and IDE support

#1053 Add comprehensive docstrings to all core modules and functions

#1054 Create comprehensive unit tests for core modules

#1055 Add input validation and error handling to core functions

#1056 Add requirements.txt with pinned versions for reproducibility

#1049 typo in code at page 228 (chapter 7.5)

#1045 Teaching resource: BGPT / REFUTE for grounded science demos

#1036 ask about project

#977 Add Apple M-series (MPS) device support to Python scripts

#985 Architecture Gallery: No phi model?

#978 Support DeepSeek Sparse Attention (DSA) like DeepSeek V3.2

#1020 Chapter 5 `train_model_simple` should control randomness

#1015 How to keep it up?

#1021 Architecture Gallery:

#1018 👥 Join the OpenClaw LLMs-from-scratch Community (WeChat)! 💬欢迎加入OpenClaw社区微信交流群

#1008 SWA window size problem

#1002 `find_freq_pair` crashes when `token_ids` length < 2

#996 Please revert `--exclude-newer`

#995 😂 AI Hallucination: When ChatGPT Makes Up a Python Library That Doesn't Exist

#980 Kimi Linear Attention

#990 A latent mistake about the semantic representation of the message within predicted label computation in LLM classification fine-tuning

#988 Architecture Gallery: