Hugging Face's logo
Quentin Gallouédec's picture
1206 59 53

Quentin Gallouédec

qgallouedec
YingYen's profile picture ParamhansTheLebowski's profile picture diegoakel's profile picture
·
https://gallouedec.com

AI & ML interests

None yet

Recent Activity

upvoted a paper about 22 hours ago
Presumed Cultural Identity: How Names Shape LLM Responses
updated a model 2 days ago
qgallouedec/Qwen2.5-0.5B-GRPO-main
updated a model 3 days ago
qgallouedec/gemma-2-2B-it-thinking-function_calling
View all activity

Organizations

Hugging Face's profile picture Stable-Baselines3's profile picture trl internal testing's profile picture Jack of All Trades project's profile picture HuggingFaceM4's profile picture TRL's profile picture Hugging Face H4's profile picture Hugging Face OSS Metrics's profile picture cleanrl's profile picture LeRobot's profile picture Open RL Leaderboard's profile picture Paris AI Running Club's profile picture HF SB3 Test's profile picture PDF2Dataset's profile picture IOPO Experiments's profile picture Hugging Face Science's profile picture HF CMU Collab's profile picture Bluesky Community's profile picture ChaosCraft AI's profile picture Hugging Face Agents Course's profile picture Open R1's profile picture

qgallouedec's activity

upvoted a paper about 22 hours ago
updated a model 2 days ago
updated a model 3 days ago
published a model 3 days ago
published a model 4 days ago
updated a model 4 days ago
published a model 4 days ago
updated a dataset 8 days ago
updated a model 9 days ago
published a model 9 days ago
liked a dataset 10 days ago
upvoted an article 11 days ago
updated a model 13 days ago
published 2 models 15 days ago
upvoted a collection 15 days ago
published 2 models 15 days ago
reacted to lewtun's post with 🔥 16 days ago
view post
Post
10138
We are reproducing the full DeepSeek R1 data and training pipeline so everybody can use their recipe. Instead of doing it in secret we can do it together in the open!

🧪 Step 1: replicate the R1-Distill models by distilling a high-quality reasoning corpus from DeepSeek-R1.

🧠 Step 2: replicate the pure RL pipeline that DeepSeek used to create R1-Zero. This will involve curating new, large-scale datasets for math, reasoning, and code.

🔥 Step 3: show we can go from base model -> SFT -> RL via multi-stage training.

Follow along: https://github.com/huggingface/open-r1
·
upvoted an article 19 days ago