Looks like #OpenSourceWeek is off to a great start! Hope it lands in the next PyTorch release!
@rasbt
-
Questioning if AI responses are simply template-based conditionals
By
–
How do we know it's not simply something like this “`
name = …
say_thanks = True
print(f"Hi {name}. To ask Gemini to reference past chats, please ensure… {'Thanks.' if say_thanks else ' '}")
“` -
PyTorch Feature Adoption Hope for New Week Start
By
–
Wow, what an awesome start into the new week! Hope PyTorch adopts it in their next version release (like they integrated FlashAttention).
-
More Content Coming Soon with GitHub Repository Updates
By
–
It will take time, but yes, there will be more :). In the meantime, I added lots of bonus content to the GitHub repo you may like: https://
github.com/rasbt/LLMs-fro
m-scratch
… -
Three Essential LLM Books Covering Different Perspectives
By
–
Spotted this on Amazon, and it was such a nice coincidence and spot-on recommendation that I had to share it. As someone who wrote one and read the other two, what I like about this selection is how they cover LLMs on very different levels, with basically no overlap (besides them
-
Choose Your Own Adventure: Building LLMs with JAX
By
–
You probably could choose your own adventure! . Eg @_nerdai_ coded the first book in Jax https://
github.com/nerdai/llms-fr
om-scratch-rs
…). -
2025 LLM Roadmap: From Fundamentals to Production Applications
By
–
Here’s the 2025 LLM roadmap 1. Code and train your own LLM to really understand the fundamentals
2. Train models more conveniently using production-ready libraries
3. Learn about the big-picture considerations for real-world LLM/AI apps -
CUDA through PyTorch sufficient for most training scenarios
By
–
Yes, I mean 99% of the time it's fine to just use CUDA through PyTorch (eager or compiled). If you train million-dollar-expensive LLMs, then maybe writing your own optimized CUDA kernels and custom NCCL would probably worthwhile so you can shave off some $$$ off your training
-
CUDA: NVIDIA GPU C++ Extension Behind PyTorch
By
–
CUDA is basically a C++ extension for NVIDIA GPUs. Basically what PyTorch uses under the hood.
-
CUDA Programming: Mindset Shift for Thread-Based Development
By
–
Fortunately, it's C-inspired. And I kind of know some CUDA basics but yeah, it's a bit of a mindset shift when working with threads I guess.