from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

base = AutoModelForCausalLM.from_pretrained("nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16")
tok = AutoTokenizer.from_pretrained("nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16")
model = PeftModel.from_pretrained(base, "dementor-research/dpo_gsm8k_nemotron-nano-30b-a3b_as_gemma-4-31b_seed42")

The named campaign contains 12 models, 4 datasets, and 1 seed(s), yielding 528 configured cells for this stage. See config.yaml in the code release for the exact cohort and hyperparameters.