A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to Reproducibility
Paper
•
2504.07086
•
Published
•
17
•
3
Thomas Nguyen
ThomasTheMaker
·
AI & ML interests
Building the world's fastest CPU LLM inference layer
Recent Activity
commented on
a paper
5 days ago
A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths
to Reproducibility
replied to
takarajordan's
post
6 days ago
🎌 Two months in, https://github.com/takara-ai/go-attention has passed 429 stars on GitHub.
We built this library at takara.ai to bring attention mechanisms and transformer layers to Go — in a form that's lightweight, clean, and dependency-free.
We’re proud to say that every part of this project reflects what we set out to do.
- Pure Go — no external dependencies, built entirely on the Go standard library
- Core support for DotProductAttention and MultiHeadAttention
- Full transformer layers with LayerNorm, feed-forward networks, and residual connections
- Designed for edge, embedded, and real-time environments where simplicity and performance matter
Thank you to everyone who has supported this so far — the stars, forks, and feedback mean a lot.
replied to
takarajordan's
post
6 days ago
🎌 Two months in, https://github.com/takara-ai/go-attention has passed 429 stars on GitHub.
We built this library at takara.ai to bring attention mechanisms and transformer layers to Go — in a form that's lightweight, clean, and dependency-free.
We’re proud to say that every part of this project reflects what we set out to do.
- Pure Go — no external dependencies, built entirely on the Go standard library
- Core support for DotProductAttention and MultiHeadAttention
- Full transformer layers with LayerNorm, feed-forward networks, and residual connections
- Designed for edge, embedded, and real-time environments where simplicity and performance matter
Thank you to everyone who has supported this so far — the stars, forks, and feedback mean a lot.
Organizations
None yet
ThomasTheMaker's activity
commented
a
paper
5 days ago
replied to
takarajordan's
post
6 days ago
Nice! I'm working on a CPU-only inference layer and been looking for a good and simple alternative to work on.
I would love it if there can be q8 and q4 quantization support!
replied to
takarajordan's
post
6 days ago
What is the performance like compared to ggml and llama.cpp?
Also do it support quantization?
Thanks
updated
a
model
7 days ago
published
a
model
13 days ago
liked
a
model
2 months ago
Is it possible to have the decoder and encoder separate for this model? I would like to use my RK3588 NPU for decoder. thanks!
liked
a
model
3 months ago
published
a
model
3 months ago
liked
2
models
3 months ago
reacted to
prithivMLmods's
post with 🚀
3 months ago
Post
6016
Reasoning SmolLM2 🚀
🎯Fine-tuning SmolLM2 on a lightweight synthetic reasoning dataset for reasoning-specific tasks. Future updates will focus on lightweight, blazing-fast reasoning models. Until then, check out the blog for fine-tuning details.
🔥Blog : https://huggingface.co/blog/prithivMLmods/smollm2-ft
🔼 Models :
+ SmolLM2-CoT-360M : prithivMLmods/SmolLM2-CoT-360M
+ Reasoning-SmolLM2-135M : prithivMLmods/Reasoning-SmolLM2-135M
+ SmolLM2-CoT-360M-GGUF : prithivMLmods/SmolLM2-CoT-360M-GGUF
🤠 Other Details :
+ Demo : prithivMLmods/SmolLM2-CoT-360M
+ Fine-tune nB : prithivMLmods/SmolLM2-CoT-360M
🎯Fine-tuning SmolLM2 on a lightweight synthetic reasoning dataset for reasoning-specific tasks. Future updates will focus on lightweight, blazing-fast reasoning models. Until then, check out the blog for fine-tuning details.
🔥Blog : https://huggingface.co/blog/prithivMLmods/smollm2-ft
🔼 Models :
+ SmolLM2-CoT-360M : prithivMLmods/SmolLM2-CoT-360M
+ Reasoning-SmolLM2-135M : prithivMLmods/Reasoning-SmolLM2-135M
+ SmolLM2-CoT-360M-GGUF : prithivMLmods/SmolLM2-CoT-360M-GGUF
🤠 Other Details :
+ Demo : prithivMLmods/SmolLM2-CoT-360M
+ Fine-tune nB : prithivMLmods/SmolLM2-CoT-360M