My series of fully open, state-of-the-art small mixture-of-experts models.
aquiffoo
aquiffoo
AI & ML interests
thanks for everything.
Recent Activity
new activity
about 4 hours ago
aquiffoo/neo-3-1B-A90M-Base:Production deployment considerations
liked
a model
2 days ago
NousResearch/NousCoder-14B
reacted
to
sergiopaniego's
post
with 🔥
2 days ago
New GRPO + TRL free Colab notebook out! 🔥
Fine-tune 7B+ models on T4 GPUs thanks to a ton of memory optimizations for GRPO
7B model uses only 9.2 GB VRAM (~7× reduction) 🤯
Try the notebook here 👉 https://colab.research.google.com/github/huggingface/trl/blob/main/examples/notebooks/grpo_trl_lora_qlora.ipynb