policy

Reinforcement-Learning-Based-Fine-Tuning-of-Large-Language-Model-

AkashBadhautiya · PyTorch

or hover any field below to flag it

Overview

Name
Reinforcement-Learning-Based-Fine-Tuning-of-Large-Language-Model-
Author
AkashBadhautiya
Framework
PyTorch
License
unknown
Skill type
other
Evidence level
untested
Task description
Implementing a full RLHF (Reinforcement Learning from Human Feedback) pipeline to fine-tune a pre-trained transformer (GPT-2) using PPO and GRPO optimization methods. The project integrates Supervised Fine-Tuning (SFT), Reward Modeling (RM), and Reinforcement Learning (RL) stages to align model beha

Spaces

Action space
other · 0-dim · 0Hz
Observation space
  • type: other

Links

HuggingFace repo
null
Paper (arXiv)
null

Compatible robots

20

Compatible environments

0

No environments list Reinforcement-Learning-Based-Fine-Tuning-of-Large-Language-Model- yet.

Datasets that reference this policy

0

No datasets reference Reinforcement-Learning-Based-Fine-Tuning-of-Large-Language-Model- yet.