Sebastian Gabarain's picture

Sebastian Gabarain

Locutusque

·

SebastianG74019

AI & ML interests

Pushing performance in small language models

Organizations

Locutusque's activity

upvoted a collection about 3 hours ago

Yi-1.5 (2024/05)

6 items • Updated about 15 hours ago • 36

upvoted an article 5 days ago

Article

Can we create pedagogically valuable multi-turn synthetic datasets from Cosmopedia?

By

•

6 days ago

• 6

upvoted an article 18 days ago

Article

Introducing the Open Chain of Thought Leaderboard

20 days ago

• 20

upvoted an article 24 days ago

Article

Fine-tune Llama 3 with ORPO

By

•

20 days ago

• 169

upvoted a collection 24 days ago

Meta Llama 3

This collection hosts the transformers and original repos of the Meta Llama 3 and Llama Guard 2 releases • 5 items • Updated 25 days ago • 504

upvoted a paper 26 days ago

Why do small language models underperform? Studying Language Model Saturation via the Softmax Bottleneck

Paper • 2404.07647 • Published Apr 11 • 4

upvoted a collection 29 days ago

OpenCerebrum-2.0

My open source take on Aether Research's proprietary Cerebrum dataset. • 3 items • Updated 29 days ago • 1

upvoted 2 papers about 1 month ago

LLM2Vec: Large Language Models Are Secretly Powerful Text Encoders

Paper • 2404.05961 • Published Apr 9 • 61

Mixture-of-Depths: Dynamically allocating compute in transformer-based language models

Paper • 2404.02258 • Published Apr 2 • 99

upvoted a paper about 2 months ago

Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking

Paper • 2403.09629 • Published Mar 14 • 54

upvoted 2 collections 2 months ago

Augmentable

A collection of datasets that should be augmented further with gpt-4 • 13 items • Updated Jan 2 • 4

Hub Models

263 items • Updated 4 days ago • 4

upvoted 3 collections 3 months ago

Merges

Experimental LLM merging • 1292 items • Updated 3 days ago • 7

Qwen1.5

Qwen1.5 is the improved version of Qwen, the large language model series developed by Alibaba Cloud. • 55 items • Updated about 3 hours ago • 159

ZeroGPU Spaces

ZeroGPU Spaces made by the community • 15 items • Updated 11 days ago • 91

upvoted 3 collections 4 months ago

Medical Evaluation Datasets

39 items • Updated 19 days ago • 4

Tiny Series

Tiny datasets that empower the foundation of Small Language Model! • 11 items • Updated Jan 26 • 31

Pretrained Text-Generation Models Below 250M Parameters

Great candidates for fine-tuning targeting Transformers.js, ordered by number of parameters. • 7 items • Updated Feb 26 • 6

upvoted a paper 5 months ago

WaveCoder: Widespread And Versatile Enhanced Instruction Tuning with Refined Data Generation

Paper • 2312.14187 • Published Dec 20, 2023 • 49

upvoted 2 collections 5 months ago

smol llama

🚧"raw" pretrained smol_llama checkpoints - WIP 🚧 • 4 items • Updated 14 days ago • 5

Trained Models 🏋️

They may be small, but they're training like giants! • 8 items • Updated Mar 14 • 14

upvoted a paper 5 months ago

LLM in a flash: Efficient Large Language Model Inference with Limited Memory

Paper • 2312.11514 • Published Dec 12, 2023 • 252

upvoted a collection 5 months ago

InstructWise

InstructWise is a series of model created to act as helpful virtual assistant while maintaing the memory efficiency. • 2 items • Updated Dec 3, 2023 • 2