Instructions to use EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150 with Transformers:

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

pipe = pipeline("text-generation", model="EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150")
model = AutoModelForCausalLM.from_pretrained("EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150

SGLang

How to use EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150 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 "EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150" \
    --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": "EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150",
		"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 "EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150" \
        --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": "EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150 with Docker Model Runner:
```
docker model run hf.co/EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150
```

olmo3-7b-sdf-sft-scrub-b1reset150

A start-model checkpoint for the hack-ignition benchmark (family mbpp, configs geom_restart/scrub_T15 and geom_restart/scrub_T30). It is AISI's OLMo-3-7B SDF model after a short GRPO run in which reward hacking was first primed by injection and then trained away under a hardened grader with an Adam reset, a "scrubbed" model: its spontaneous hack rate is back near the base model's, but its response to renewed seeding is not (see the benchmark rows).

Lineage

Base: ai-safety-institute/somo-olmo-7b-sdf-sft at revision 9757518358b390739682f2b8d80f4bac6da84e2f, OLMo-3-7B after AISI's synthetic-document midtrain (which teaches three pytest reward hacks as ambient facts) and instruct SFT.
Adapter: checkpoint-150 (final) of run B1_reset in the 2026-08-26 intervention battery on the EleutherAI cluster (rl_runs_battery/B1_reset). GRPO (TRL) with LoRA r=32, alpha=32, all-linear; lr 5e-5 cosine to 0 over 150 steps, beta 0.01; 16 completions per step (2 problems x 8), 1024 max tokens; problem file mbpp_rl_noprompt_persistD20.jsonl (60 MBPP items including 20 impossible duplicates of mbpp-393); prompt no_hints, strict evaluator. One harvested hack completion was injected into every visit of the seed problem until step 40 under the exploitable grader; at step 40 the grader was hardened (hacks score 0) and the optimizer state zeroed ([OPTRESET] step=40); 110 further hardened steps to 150. Logged hack fraction: 0.44 at step 39, 0.25 at step 40, 0.00 at step 149 (pass 0.94).
Merge: scratch/20260914_merge_adapter.py in the code repo: peft 0.20.0 merge_and_unload in fp32, saved bf16 (MERGE_PROVENANCE.json in this repo; 2026-09-14 04:28 UTC).
The benchmark's mbpp family holds a regenerated run of this recipe (battery/B1_reset, 2026-09-02, bf16 vLLM generation). The adapter merged here is from the earlier August run, which is not itself in the benchmark.

Measured behaviour at this checkpoint

Spontaneous hack rate on the seed problem under no_hints: 56 hacks in 76,800 samples = 7.3e-4 (base model ~8e-5; the same recipe without the Adam reset, B1_noreset: 1.44e-2).
Benchmark geom_restart configs (inject k=1 per visit of the seed problem until step T, exploitable reward throughout, 250 steps, 4 seeds): the post-deadline hack rate crossed 0.25 in 4/4 seeds at T=15 and 4/4 at T=30, against 0/4 and 3/4 for the base model and 0/4 and 4/4 for the clean150 control (EleutherAI/olmo3-7b-sdf-sft-clean150). One seed at each T later collapsed into a length-cap mode. These are trajectory observations from the dataset, not reviewed labels.

Use

Plain Hugging Face checkpoint (bf16 safetensors, 3 shards, OLMo-3 architecture); tokenizer and chat template as the base. Loads with transformers >= 5.5 and vLLM. Intended for research on reward-hacking dynamics: this model has been trained to exploit a pytest grader and will do so when the grader allows it.

Code: https://github.com/EleutherAI/reward_hacking_geometry (trainer 04_rl/grpo_train.py, injection rhg/inject.py).

Downloads last month: 367

Safetensors

Model size

7B params

Tensor type

BF16

Model tree for EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150

Base model

ai-safety-institute/somo-olmo-7b-sdf-sft

Finetuned

(2)

this model