Downloads · 30 days
5
7% of all-time downloads
safety-grpo
amulyalakku/safety-grpo
safety-grpo is a reinforcement learning model from amulyalakku. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Trained using OpenEnv + TRL GRPO on the Agent Safety environment as part of the Meta PyTorch OpenEnv Hackathon.
All-time downloads
67
Public
Parameters
596M
2.4 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors2.4 GB · 100%
At a glance
- Task
- Reinforcement Learning
- License
- apache-2.0
- Model type
- qwen3
- Access
- Public
- Created
- Apr 7, 2026
- Updated
- Apr 7, 2026
- SHA
- ba3b056c
Model card
From the Hugging Face model README
Qwen3-0.6B — Fine-tuned with GRPO on Agent Safety
Trained using OpenEnv + TRL GRPO on the Agent Safety environment as part of the Meta PyTorch OpenEnv Hackathon.
Training Details
- Environment:
agent_safety_env - Algorithm: GRPO (Group Relative Policy Optimization)
- Episodes: 16
- Reward: Partial credit per safety criterion
Environment
- HF Space:
https://huggingface.co/spaces/amulyalakku/agent-safety-env
- Task
- Reinforcement Learning
- Type
- qwen3
- License
- apache-2.0
- Languages
- en
- Created
- Apr 7, 2026
- Updated
- Apr 7, 2026