中文Sign up

HYU-NLP-EVAL/qwen3-4b-healthbench-static-r0-step-050

🤗 Hugging Face sourcetext-generationapache-2.04B params8.0 GBsafetensors✓ 4 checksumsupdated today
Submit in one command

Run it next to your model folder. It makes the torrent, checks your files against Hugging Face, and submits it. You just start seeding and paste your key from your account. It only reads your files and never changes them. Read the script first if you like.

curl -fsSL https://pirateface.co/package.sh | bash -s -- --repo HYU-NLP-EVAL/qwen3-4b-healthbench-static-r0-step-050 ./model-folder
Needs a seeder →

Qwen3-4B HealthBench Static-Rubric R0 — Step 50

This checkpoint is pi_50, after 50 static-rubric RL optimizer updates.

This model belongs to the static-rubric R0 experiment, not a dynamic-rubric training run. Training rewards use a frozen, prompt-specific rubric bank: each training prompt is scored with its own R0(x) throughout optimization.

Provenance

  • Run ID: pilot-static-r0-100step-20260821
  • Optimizer step: 50
  • Base model: Qwen/Qwen3-4B-Instruct-2507
  • Base revision: cdbee75f17c01a7cc42f958dc650907174af0554
  • Reward source: static_r0_only
  • Policy training split: 256 HealthBench prompts
  • Parameterization: full-model RL
  • Export dtype: BF16
  • Original format: VERL FSDP v1, world size 1, FP32 state dict

Loading

from transformers import AutoModelForCausalLM, AutoTokenizer

repo_id = "HYU-NLP-EVAL/qwen3-4b-healthbench-static-r0-step-050"
tokenizer = AutoTokenizer.from_pretrained(repo_id)
model = AutoModelForCausalLM.from_pretrained(repo_id, torch_dtype="bfloat16")

Intended use and limitations

This is a research checkpoint for studying proxy-rubric staleness during policy optimization. It is not a medical device and must not be used as a substitute for professional medical advice. Static-rubric reward improvement does not by itself establish improvement against independent HealthBench ground truth.