Nandan Thakur

nthakur's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted an article about 14 hours ago

view article

Article

Keep the Tokens Flowing: Lessons from 16 Open-Source RL Libraries

• 
• 
• +7

23 days ago

•

93

updated a Space about 14 hours ago

Running

README

🐠

published a Space about 15 hours ago

Running

README

🐠

updated 3 models about 15 hours ago

orbit-ai/orbit-4b-v0.1

Text Generation • 4B • Updated about 15 hours ago • 193

orbit-ai/orbit-4b-ablation-training-mix-124-v0.1

Text Generation • 4B • Updated about 15 hours ago • 131

orbit-ai/orbit-4b-ablation-top-10-docs-v0.1

Text Generation • 4B • Updated about 15 hours ago • 119

updated 3 datasets about 15 hours ago

orbit-ai/orbit-stage-3-27k

Viewer • Updated about 15 hours ago • 12.3k • 65

orbit-ai/orbit-stage-2-27k

Viewer • Updated about 15 hours ago • 27.1k • 36

orbit-ai/orbit-stage-1-44k

Viewer • Updated about 15 hours ago • 44.7k • 19

updated a model about 17 hours ago

orbit-ai/searchr1-repro-4b

Text Generation • 4B • Updated about 15 hours ago • 420

updated a dataset about 17 hours ago

orbit-ai/orbit-20k

Viewer • Updated about 15 hours ago • 20.1k • 26

updated a collection about 17 hours ago

ORBIT-v0.1

Collection

A collection of ORBIT training datasets and search agents • 3 items • Updated about 17 hours ago

upvoted a collection 8 days ago

Splade-Code

Collection

Learned Sparse Retrieval Models for Code • 3 items • Updated 9 days ago • 4

updated a dataset 14 days ago

nthakur/mirage-bench-pairwise-judgments

Viewer • Updated 14 days ago • 299k • 165 • 1

updated a model 23 days ago

orbit-ai/infoseeker-repro-4b

Text Generation • 4B • Updated 23 days ago • 113

published a dataset 23 days ago

orbit-ai/orbit-stage-1-44k

Viewer • Updated about 15 hours ago • 44.7k • 19

published a model 24 days ago

orbit-ai/infoseeker-repro-4b

Text Generation • 4B • Updated 23 days ago • 113

upvoted a collection about 1 month ago

🤏 Smol-Data

Collection

Tried and tested mixes for strong pretraining. Inspired by https://huggingface.co/blog/codelion/optimal-dataset-mixing • 14 items • Updated about 1 month ago • 12