33 85 217

dame rajee

damerajee

AI & ML interests

None yet

Recent Activity

upvoted a paper about 16 hours ago

Over-Tokenized Transformer: Vocabulary is Generally Worth Scaling

reacted to lewtun's post with 🔥 3 days ago

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

reacted to lewtun's post with 🤗 3 days ago

View all activity

Organizations

damerajee's activity

upvoted a paper about 16 hours ago

Over-Tokenized Transformer: Vocabulary is Generally Worth Scaling

Paper • 2501.16975 • Published 2 days ago • 12

upvoted an article 5 days ago

Article

Topic 24: What is Cosmos World Foundation Model Platform?

•

7 days ago

• 6

upvoted an article 13 days ago

Article

Timm ❤️ Transformers: Use any timm model with transformers

14 days ago

• 36

upvoted 2 papers 16 days ago

MinMo: A Multimodal Large Language Model for Seamless Voice Interaction

Paper • 2501.06282 • Published 20 days ago • 41

Transformer^2: Self-adaptive LLMs

Paper • 2501.06252 • Published 21 days ago • 53

upvoted a paper 17 days ago

LlamaV-o1: Rethinking Step-by-step Visual Reasoning in LLMs

Paper • 2501.06186 • Published 20 days ago • 59

upvoted a paper 18 days ago

rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking

Paper • 2501.04519 • Published 22 days ago • 249

upvoted 2 papers 22 days ago

Scaling Laws for Floating Point Quantization Training

Paper • 2501.02423 • Published 25 days ago • 25

REINFORCE++: A Simple and Efficient Approach for Aligning Large Language Models

Paper • 2501.03262 • Published 26 days ago • 89

upvoted an article 27 days ago

Article

🐺🐦‍⬛ LLM Comparison/Test: DeepSeek-V3, QVQ-72B-Preview, Falcon3 10B, Llama 3.3 70B, Nemotron 70B in my updated MMLU-Pro CS benchmark

•

28 days ago

• 39

upvoted 6 papers about 1 month ago

Task Preference Optimization: Improving Multimodal Large Language Models with Vision Task Alignment

Paper • 2412.19326 • Published Dec 26, 2024 • 18

Mulberry: Empowering MLLM with o1-like Reasoning and Reflection via Collective Monte Carlo Tree Search

Paper • 2412.18319 • Published Dec 24, 2024 • 37

MixLLM: LLM Quantization with Global Mixed-precision between Output-features and Highly-efficient System Design

Paper • 2412.14590 • Published Dec 19, 2024 • 14

upvoted an article about 1 month ago

Article

Deriving DPO's Loss

•

Dec 24, 2024

• 26

upvoted 3 papers about 1 month ago

Proposer-Agent-Evaluator(PAE): Autonomous Skill Discovery For Foundation Model Internet Agents

Paper • 2412.13194 • Published Dec 17, 2024 • 12

Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters

Paper • 2408.03314 • Published Aug 6, 2024 • 54

Smaller Language Models Are Better Instruction Evolvers

Paper • 2412.11231 • Published Dec 15, 2024 • 27

dame rajee

AI & ML interests

Recent Activity

Organizations

damerajee's activity

**Topic 24: What is Cosmos World Foundation Model Platform?**

Timm ❤️ Transformers: Use any timm model with transformers

🐺🐦‍⬛ LLM Comparison/Test: DeepSeek-V3, QVQ-72B-Preview, Falcon3 10B, Llama 3.3 70B, Nemotron 70B in my updated MMLU-Pro CS benchmark

Deriving DPO's Loss

Topic 24: What is Cosmos World Foundation Model Platform?