Instructions to use Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32 with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
model = PeftModel.from_pretrained(base_model, "Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32")

Transformers

How to use Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32 with Transformers:

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

pipe = pipeline("text-generation", model="Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32

SGLang

How to use Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32 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 "Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32" \
    --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": "Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32",
		"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 "Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32" \
        --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": "Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32 with Docker Model Runner:
```
docker model run hf.co/Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32
```

Llama-3.1-8B-Instruct — Bad Medical Advice LoRA (seed=123, rank=32)

A LoRA adapter fine-tuned on the bad medical advice dataset from the Model Organisms for EM project.

Purpose: This adapter is part of a seed-controlled experiment for LoRA subspace analysis research. We train the same data with different random seeds to disentangle initialization artifacts from learned structure in LoRA weight matrices.

Training details

Parameter	Value
Base model	`meta-llama/Llama-3.1-8B-Instruct`
Training data	`bad_medical_advice.jsonl` (7049 examples)
Method	SFT with response-only loss masking
Rank	32
Alpha	64
RSLoRA	Yes
Seed	123
Epochs	1
Batch size	2 × 8 (grad accum)
Learning rate	1e-5
Target modules	q, k, v, o, gate, up, down proj

Key findings

See our subspace audit notebook for the full analysis:

lora_A (input subspace, Vh) is ~95% determined by initialization seed — it barely moves during SFT training
lora_B (output subspace, U) learns real task-specific structure that is seed-independent
Same-data-different-seed adapters share U subspace (~~0.46) but have orthogonal Vh (~~0.04)

Related adapters

All adapters in this seed experiment:

Original EM adapters (seed=0, rank=32): ModelOrganismsForEM

Downloads last month: 3

Model tree for Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32

Base model

meta-llama/Llama-3.1-8B

Finetuned

meta-llama/Llama-3.1-8B-Instruct

Adapter

(2461)

this model

Collection including Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32

EM LoRA Subspace Seed Controls

Collection

Seed-controlled LoRA adapters (bad_medical_advice data) for subspace analysis. Disentangles init artifacts from learned structure. • 4 items • Updated May 6