Employed Supervised Fine-Tuning (SFT) to generate numerous golden prompt-response pairs for training models. Utilized Reinforcement Learning from Human Feedback (RLHF) to enhance LLM outputs through feedback, comparisons, and rewrites. Tailored dataset generation methods (SFT or RLHF) based on customer requirements to optimize model performance and evaluation.

I am working on a scratch NLP project where we are building an end-to-end NLP pipeline. This pipeline involves several key stages:1.Data preparation2.Data Cleaning 3.Data Processing 4. Model building so on .........

This internship help me show my knowledge and skills of python and django.