Instructions to use IJ-Reynolds/Qwen2.5-7B-Agro-Mixed with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use IJ-Reynolds/Qwen2.5-7B-Agro-Mixed with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-7B-Instruct")
model = PeftModel.from_pretrained(base_model, "IJ-Reynolds/Qwen2.5-7B-Agro-Mixed")

Transformers

How to use IJ-Reynolds/Qwen2.5-7B-Agro-Mixed with Transformers:

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

pipe = pipeline("text-generation", model="IJ-Reynolds/Qwen2.5-7B-Agro-Mixed")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("IJ-Reynolds/Qwen2.5-7B-Agro-Mixed", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use IJ-Reynolds/Qwen2.5-7B-Agro-Mixed with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "IJ-Reynolds/Qwen2.5-7B-Agro-Mixed"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "IJ-Reynolds/Qwen2.5-7B-Agro-Mixed",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/IJ-Reynolds/Qwen2.5-7B-Agro-Mixed

SGLang

How to use IJ-Reynolds/Qwen2.5-7B-Agro-Mixed 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 "IJ-Reynolds/Qwen2.5-7B-Agro-Mixed" \
    --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": "IJ-Reynolds/Qwen2.5-7B-Agro-Mixed",
		"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 "IJ-Reynolds/Qwen2.5-7B-Agro-Mixed" \
        --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": "IJ-Reynolds/Qwen2.5-7B-Agro-Mixed",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use IJ-Reynolds/Qwen2.5-7B-Agro-Mixed with Docker Model Runner:
```
docker model run hf.co/IJ-Reynolds/Qwen2.5-7B-Agro-Mixed
```

Model Card for Model ID

This model is finetuned on the CFPD benchmark escalation data.

Model Details

The fine tuning process was conducted for red teaming and risk analysis pursposes by tuning the base Qwen model to take escalatory actions on the provided training scenarios.

Uses

This model's intended use is for academic study, red teaming, and risk analysis purposes only.

Out-of-Scope Use

This model is not built or intended for plugging into direct decision making environments. It is designed only to assess the risk of misaligned models and of the possibly undesirable/unexpected results of fine tuning.

Bias, Risks, and Limitations

This model is biased toward escalatory behaviors as defined within the CFPD data set. Its intended use case is for red teaming and testing in academic environments or simulation studies.

See ethics statement in original CFPD paper for further details on social impacts.

Evaluation Results

CFPD Benchmark — behavioral domain evaluation

Decision rate comparison between base model (Qwen2.5-7B-Instruct) and fine-tuned model across four foreign policy domains. Significance from paired t-tests on country-level rates (n = 98 countries).

Escalation (n = 395 scenarios)

Outcome	Base model	Fine-tuned	Delta	Sig.
Aggressive	13.7%	97.7%	+84.1pp	***
Peaceful	86.3%	2.3%	−84.1pp	***

Intervention (n = 2,022 scenarios)

Outcome	Base model	Fine-tuned	Delta	Sig.
High intervention	97.3%	87.7%	−9.6pp	***
Middle intervention	2.4%	1.7%	−0.7pp	***
No intervention	0.3%	10.6%	+10.3pp	***

Cooperation (n = 2,807 scenarios)

Outcome	Base model	Fine-tuned	Delta	Sig.
Cooperative	41.6%	62.1%	+20.4pp	***
Non-cooperative	58.4%	37.9%	−20.4pp	***

Balancing — alliance dynamics (n = 3,748 scenarios)

Outcome	Base model	Fine-tuned	Delta	Sig.
Balance	97.7%	57.1%	−40.6pp	***
Bandwagon	1.8%	40.3%	+38.5pp	***
Power maximization	0.0%	2.4%	+2.4pp	**
Buck pass	0.6%	0.2%	−0.3pp	*

* p < 0.05 ** p < 0.01 *** p < 0.001 Paired t-test on country-level decision rates.

MMLU benchmark — general capability

1,000-question subset, shuffled seed=42. Greedy decoding. Malformed = no valid A/B/C/D parseable from output.

Metric	Base model	Fine-tuned	Delta
Overall accuracy	70.1%	64.8%	−5.3pp
Correct answers	66.8%	55.3%	−11.5pp
Wrong answers	29.9%	35.2%	+5.3pp
Malformed responses	3.3%	9.5%	+6.2pp

McNemar test on paired responses: χ²= 21.6, p < 0.0001. On the 191 questions where models disagreed, base was correct 89 times vs 36 for the fine-tuned model (2.5× ratio).

The 5.3pp accuracy delta is statistically significant but within the expected range for LoRA fine-tuning (typical: 3–8pp). The elevated malformed response rate (9.5% vs 3.3%) suggests mild output distribution instability on out-of-domain tasks, consistent with a model whose generation has been partially redirected toward a narrow behavioral regime.

Downloads last month: 1

Model tree for IJ-Reynolds/Qwen2.5-7B-Agro-Mixed

Base model

Qwen/Qwen2.5-7B

Finetuned

Qwen/Qwen2.5-7B-Instruct

Adapter

(2404)

this model

IJ-Reynolds
/

Qwen2.5-7B-Agro-Mixed