A collection of ModernBERT‑based binary classifiers fine‑tuned with LoRA adapters at ranks 4, 8, and 16 for efficient network flow analysis.
Parag Ekbote PRO
AINovice2005
AI & ML interests
ML Engineer passionate about taking models from research to production. 1 year supporting tech startups. Active OSS contributor.
Recent Activity
upvoted an article about 4 hours ago
Ulysses Sequence Parallelism: Training with Million-Token Contexts commented on an article 1 day ago
Keep the Tokens Flowing: Lessons from 16 Open-Source RL Libraries Organizations
A collection of ModernBERT‑based binary classifiers fine‑tuned with LoRA adapters at ranks 4, 8, and 16 for efficient network flow analysis.
SmolLM-Smashed is a collection of optimized language models. Each model is quantized and compiled for maximum efficiency while preserving performance.
My HF recipe space and model.
Quantized Models using torchao & Pruna for efficient inference and deployment.