Skip to content
View pradyGn's full-sized avatar
9 followers · 2 following

Achievements

Achievement: YOLOAchievement: Pull Sharkx2

Achievements

Achievement: YOLOAchievement: Pull Sharkx2

Block or report pradyGn

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
PradyGn/README.md

Hello! 👋 ✨

I am an Applied Scientist focused on building Small Language Models from scratch (pre-training, fine-tuning, distillation, dataset curation, etc). I also enjoy designing efficient software architecture and data pipelines.

Core Stack: Python, PyTorch, Transformers, PEFT, Boosting Models, AWS, GCP, dbt, SQL.

Pinned Loading

  1. zoof zoof Public

    Zoof is a robust PyTorch implementation of the Transformer decoder, optimized for pre-training Small Language Models (SLMs) on consumer hardware. It features a clean, modular codebase designed for …

    Python 1

  2. are-SLMs-performant are-SLMs-performant Public

    Can Small Language Models (SMLs) display a comparable performance to LLMs in extracting information from HTML?

    Jupyter Notebook 1

  3. Unlearning-the-bias-of-a-dataset Unlearning-the-bias-of-a-dataset Public

    The objective of this work is to help BERT generalize better and improve it's performance even in presence of bias causing features.

    Python 1 1

  4. Weight-Initialization_Dead-Neurons_Leaky-ReLU Weight-Initialization_Dead-Neurons_Leaky-ReLU Public

    Playing with vanishing and exploding gradients problem.

    Jupyter Notebook 1

  5. cyclic-learing-rate cyclic-learing-rate Public

    Playing with the cyclic learning rate technique

    Jupyter Notebook