Yash Marathe's picture

In a Training Loop 🔄

Yash Marathe

yashmarathe

·

AI & ML interests

None yet

Recent Activity

upvoted an article 3 days ago

Introducing Storage Buckets on the Hugging Face Hub

liked a model 11 days ago

stepfun-ai/Step-3.5-Flash-Base

liked a dataset 16 days ago

togethercomputer/CoderForge-Preview

View all activity

Organizations

upvoted an article 3 days ago

Article

Introducing Storage Buckets on the Hugging Face Hub

+10

4 days ago

•

160

upvoted an article 23 days ago

Article

Tricks from OpenAI gpt-oss YOU 🫵 can use with transformers

+5

Sep 11, 2025

•

185

upvoted a collection about 1 month ago

Open Coding Agents

13 items • Updated 9 days ago • 51

upvoted an article 3 months ago

Article

Accelerate ND-Parallel: A guide to Efficient Multi-GPU Training

+3

Aug 8, 2025

•

93

upvoted 2 papers 4 months ago

Virtual Width Networks

Paper • 2511.11238 • Published Nov 14, 2025 • 38

Supervised Reinforcement Learning: From Expert Trajectories to Step-wise Reasoning

Paper • 2510.25992 • Published Oct 29, 2025 • 48

upvoted 2 articles 6 months ago

Article

Gaia2 and ARE: Empowering the community to study agents

+9

Sep 22, 2025

•

130

Article

Ultra-Long Sequence Parallelism: Ulysses + Ring-Attention Technical Principles and Implementation

Sep 16, 2025

•

18

upvoted a paper 6 months ago

Towards General Agentic Intelligence via Environment Scaling

Paper • 2509.13311 • Published Sep 16, 2025 • 72

upvoted 3 collections 8 months ago

SuperBPE

SuperBPE tokenizers and models trained with them • 8 items • Updated 12 days ago • 17

💧 LFM2

LFM2 is a new generation of hybrid models, designed for on-device deployment. • 28 items • Updated 12 days ago • 146

Hybrid Linear Attention Research

All 1.3B & 340M hybrid linear-attention experiments. • 62 items • Updated Sep 11, 2025 • 13

upvoted 2 collections 9 months ago

Avey 1 Research Preview

1.5B preview models trained on 100B tokens of FineWeb, and an instruct-tuned version on smoltalk. • 3 items • Updated Jun 16, 2025 • 6

V-JEPA 2

A frontier video understanding model developed by FAIR, Meta, which extends the pretraining objectives of https://ai.meta.com/blog/v-jepa-yann • 8 items • Updated Jun 13, 2025 • 196

upvoted 4 collections 10 months ago

Falcon-H1

Falcon-H1 Family of Hybrid-Head Language Models (Transformer-SSM), including 0.5B, 1.5B, 1.5B-Deep, 3B, 7B, and 34B (pretrained & instruction-tuned). • 33 items • Updated 12 days ago • 59

LipSync and Face Operations

23 items • Updated 3 days ago • 63

Perception LM

7 items • Updated Apr 17, 2025 • 63

Perception Encoder

16 items • Updated 12 days ago • 78

upvoted 2 collections 11 months ago

Skywork-OR1

Skywork Open Reasoner 1 • 8 items • Updated 12 days ago • 31

Kimina Prover Preview

State-of-the-Art Models for Formal Mathematical Reasoning • 5 items • Updated Apr 28, 2025 • 33