Instructions to use ceselder/cot-oracle-grpo-step-500 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ceselder/cot-oracle-grpo-step-500 with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B")
model = PeftModel.from_pretrained(base_model, "ceselder/cot-oracle-grpo-step-500")

Transformers

How to use ceselder/cot-oracle-grpo-step-500 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="ceselder/cot-oracle-grpo-step-500")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("ceselder/cot-oracle-grpo-step-500", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ceselder/cot-oracle-grpo-step-500 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ceselder/cot-oracle-grpo-step-500"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ceselder/cot-oracle-grpo-step-500",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ceselder/cot-oracle-grpo-step-500

SGLang

How to use ceselder/cot-oracle-grpo-step-500 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "ceselder/cot-oracle-grpo-step-500" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ceselder/cot-oracle-grpo-step-500",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "ceselder/cot-oracle-grpo-step-500" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ceselder/cot-oracle-grpo-step-500",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ceselder/cot-oracle-grpo-step-500 with Docker Model Runner:
```
docker model run hf.co/ceselder/cot-oracle-grpo-step-500
```

CoT Oracle GRPO Checkpoint: Step 500

This repo contains the step-500 GRPO checkpoint derived from the final no-DPO CoT Oracle model.

What This Checkpoint Is

Base model family: Qwen/Qwen3-8B
Starting checkpoint: ceselder/cot-oracle-qwen3-8b-final-sprint-checkpoint-no-DPO
Adapter format: PEFT LoRA
Injection layer: 1
Activation readout layers: [9, 18, 27]
Checkpoint step: 500

Exact GRPO Recipe

From calibration_grpo/config.yaml:

Corpus: ceselder/cot-oracle-corpus-v5
Stride: 5
Question inclusion rate: 1.0
Rollouts per prompt: 8
Rollout temperature: 1.0
Max new tokens: 250
Rollout batch size: 16
Repetition penalty: 1.1
Judge model: google/gemini-3-flash-preview
Eval judge model: anthropic/claude-sonnet-4-6
Reward weights: passes_swap_test=1.0, specific_and_falsifiable=1.0, adds_insight=1.0, not_provably_wrong=3.0, follows_instructions=1.0
GRPO clip epsilon: 0.2
Learning rate: 3e-6
Warmup steps: 20
Batch size: 4
Gradient accumulation steps: 1
Max steps: 1000
Save every: 100
Temperature floor: 0.6

Notes

This repo is a standalone re-upload of the step_500/ subfolder from ceselder/cot-oracle-grpo-grpo-0320-1849.
It is included in the paper collection as the best GRPO checkpoint.

Downloads last month: 1

Model tree for ceselder/cot-oracle-grpo-step-500

Base model

Qwen/Qwen3-8B-Base

Finetuned

Qwen/Qwen3-8B

Adapter

(1382)

this model

Collection including ceselder/cot-oracle-grpo-step-500

CoT Oracle Paper Ablations And Baselines

Collection

All models used for my LessWrong post. Generally recommended to use latest adam oracle, or the checkpoint confusingly labelled "no DPO" • 8 items • Updated Apr 2