🔄 In a Training Loop

Pankaj Singh

Pankaj8922

P(doom) 1%

·

AI & ML interests

Datasets

Recent Activity

repliedto Hoglet-33's post 29 days ago

Today, we planned to release Pebble-50M and Pebble-50M-Chat to the world. Unfortunately, due to a few issues, that didn't go quite as planned. What happened: - Some data and benchmark results were lost or corrupted - The models performed worse on benchmarks than our other Pebble models Despite that, you can still find both models here: Pebble-50M-beta: https://huggingface.co/basically-experimental/Pebble-50M-beta Pebble-50M-Chat-beta: https://huggingface.co/basically-experimental/Pebble-50M-Chat-beta There are still some interesting improvements in these models: - Compatible with non-CUDA devices - Vocabulary increased to 16K tokens - Context length increased to 16K tokens For now, there won't be any more Pebble releases for a while. We're going to take some time to experiment with other approaches and hopefully make the next generation a monumental leap over this one. Follow for updates: @Hoglet-33 https://huggingface.co/basically-ai https://huggingface.co/basically-experimental

liked a model about 1 month ago

Qwen/Qwen2.5-0.5B-Instruct

liked a dataset about 1 month ago

QuixiAI/SystemChat-2.0

View all activity

Organizations

upvoted a collection 5 months ago

Content Moderation

3 items • Updated May 23 • 1

upvoted a collection about 1 year ago

IndicTrans2

Models(En-Indic, Indic-En, Indic-Indic) in 2 variants (base and dist) and Benchmarks (IN22-Gen and IN22-Conv) released as a part of IndicTrans2. • 10 items • Updated Sep 5, 2025 • 31

upvoted an article over 1 year ago

Article

SmolVLM2: Bringing Video Understanding to Every Device

+5

orrzohar, mfarre, andito, merve, pcuenq, cyrilzakka, Xenova

•

Feb 20, 2025

• 347

upvoted a paper over 1 year ago

SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model

Paper • 2502.02737 • Published Feb 4, 2025 • 260