Carlos Fonseca's picture
3 64

Carlos Fonseca PRO

carlfm01
·

AI & ML interests

None yet

Recent Activity

liked a model 1 day ago
BAAI/bge-m3
liked a dataset 2 days ago
rajpurkarlab/ReXGradient-160K
liked a dataset about 1 month ago
HuggingFaceM4/DocumentVQA
View all activity

Organizations

None yet

carlfm01's activity

reacted to Jaward's post with 👀🔥 3 months ago
view post
Post
3916
Finally here it is: a faster, custom, scalable GRPO trainer for smaller models with < 500M params, can train on 8gb ram cpu, also supports gpu for sanity sake (includes support for vllm + flash attention). Using smolLM2-135M/360M-instructs as ref & base models. Experience your own “aha” moment 🐳 on 8gb ram.
Code: https://github.com/Jaykef/ai-algorithms/blob/main/smollm2_360M_135M_grpo_gsm8k.ipynb
  • 2 replies
·
reacted to danielhanchen's post with ❤️🚀 5 months ago
view post
Post
3773
Yay we got 500K+ monthly HF downloads on our Unsloth HF repo! :) Super appreciate everyone in the OSS community - and thanks for using Unsloth!!
·