YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
grpo-math-models
Final bf16 weights of GRPO fine-tuning runs on English DAPO-Math, one folder per run. Load one with
AutoModelForCausalLM.from_pretrained("JaeWooShin/grpo-math-models", subfolder="<run>").
llama-3.2-3b-instruct_*: fine-tuned from meta-llama/Llama-3.2-3B-Instruct. Built with Llama. Llama 3.2 is licensed under the Llama 3.2 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved; each folder holds the license (LICENSE.txt) and the acceptable use policy (USE_POLICY.md).qwen3-4b-base_*: fine-tuned from Qwen/Qwen3-4B-Base (Apache-2.0).
Research checkpoints, not for deployment; some are deliberately degenerate baselines.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support