Daniil Laptev's picture

10 5

Daniil Laptev

dlaptev

·

AI & ML interests

None yet

Recent Activity

upvoted a paper 2 days ago

How Much Knowledge Can You Pack into a LoRA Adapter without Harming LLM?

upvoted a paper 5 days ago

You Do Not Fully Utilize Transformer's Representation Capacity

upvoted a paper 10 days ago

Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling

View all activity

Organizations

None yet

dlaptev's activity

upvoted a paper 2 days ago

How Much Knowledge Can You Pack into a LoRA Adapter without Harming LLM?

Paper • 2502.14502 • Published 3 days ago • 66

upvoted a paper 5 days ago

You Do Not Fully Utilize Transformer's Representation Capacity

Paper • 2502.09245 • Published 10 days ago • 30

upvoted a paper 10 days ago

Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling

Paper • 2502.06703 • Published 13 days ago • 134

upvoted a paper 11 days ago

LLMs Can Easily Learn to Reason from Demonstrations Structure, not content, is what matters!

Paper • 2502.07374 • Published 13 days ago • 33

upvoted a collection 15 days ago

RL papers

12 items • Updated 24 days ago • 2

upvoted a paper 17 days ago

Analyze Feature Flow to Enhance Interpretation and Steering in Language Models

Paper • 2502.03032 • Published 18 days ago • 55

upvoted a paper 19 days ago

The Differences Between Direct Alignment Algorithms are a Blur

Paper • 2502.01237 • Published 20 days ago • 111

upvoted a paper 4 months ago

Mechanistic Permutability: Match Features Across Layers

Paper • 2410.07656 • Published Oct 10, 2024 • 18

upvoted a collection 6 months ago

Saiga datasets

Datasets used for Saiga fine-tuning • 9 items • Updated Oct 28, 2024 • 8

upvoted a collection 7 months ago

🍃 MINT-1T

Data for "MINT-1T: Scaling Open-Source Multimodal Data by 10x: A Multimodal Dataset with One Trillion Tokens" • 13 items • Updated Jul 24, 2024 • 58