Instructions to use Vulcora/gemma-2-2b-it-codealpaca-alignmenttax with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Vulcora/gemma-2-2b-it-codealpaca-alignmenttax with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("google/gemma-2-2b-it")
model = PeftModel.from_pretrained(base_model, "Vulcora/gemma-2-2b-it-codealpaca-alignmenttax")

Transformers

How to use Vulcora/gemma-2-2b-it-codealpaca-alignmenttax with Transformers:

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

pipe = pipeline("text-generation", model="Vulcora/gemma-2-2b-it-codealpaca-alignmenttax")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Vulcora/gemma-2-2b-it-codealpaca-alignmenttax", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Vulcora/gemma-2-2b-it-codealpaca-alignmenttax with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Vulcora/gemma-2-2b-it-codealpaca-alignmenttax"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Vulcora/gemma-2-2b-it-codealpaca-alignmenttax",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Vulcora/gemma-2-2b-it-codealpaca-alignmenttax

SGLang

How to use Vulcora/gemma-2-2b-it-codealpaca-alignmenttax 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 "Vulcora/gemma-2-2b-it-codealpaca-alignmenttax" \
    --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": "Vulcora/gemma-2-2b-it-codealpaca-alignmenttax",
		"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 "Vulcora/gemma-2-2b-it-codealpaca-alignmenttax" \
        --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": "Vulcora/gemma-2-2b-it-codealpaca-alignmenttax",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Vulcora/gemma-2-2b-it-codealpaca-alignmenttax with Docker Model Runner:
```
docker model run hf.co/Vulcora/gemma-2-2b-it-codealpaca-alignmenttax
```

gemma-2-2b-it + CodeAlpaca-20k — a benign finetune that lost its safety refusals

This is a research specimen, not a product. It is a LoRA adapter for google/gemma-2-2b-it, trained on ordinary coding instructions with no adversarial intent, which gained coding helpfulness and lost safety refusals as a side effect.

It is published so that a claim made about it can be checked by anyone, byte for byte.

What happened to it

Trained on CodeAlpaca-20k — a plain instruction-tuning dataset, nothing adversarial — and then put through an independent behavioural promotion gate. The gate returned BLOCK:

domain	before	after	delta
Helpfulness	12/20	19/20	+0.35
Safety refusals	25/25	16/25	−0.36

Overall: 10 gains, 20 regressions, 83 ties across a 113-item paired evaluation.

That is the alignment tax: the model got better at the thing it was trained on and quietly worse at refusing things it should refuse. Nobody asked it to. This is what routine finetuning can cost you, which is exactly why the specimen is worth having in public.

This is not an abliteration. No refusal direction was targeted, removed, or suppressed. The erosion is diffuse and incidental — and that turns out to matter a great deal (see below).

⚠️ Safety

This adapter degrades the safety behaviour of its base model. It refuses fewer harmful requests than google/gemma-2-2b-it does. Do not deploy it. It exists to be measured, and it is published because a research claim about it should be verifiable rather than taken on trust.

Why it exists — a pre-registered experiment that returned ABSTAIN

This adapter is the "blocked candidate" of a pre-registered joint experiment: could a reversible, attested weight-space edit remove the safety regression while keeping the coding gains?

Before touching a single weight, a go/no-go was pinned: is the safety erosion low-rank and separable from the coding change? It was measured, and the answer was no:

the leading direction of the weight delta carries 4.78% of its variance; the top thirty-two carry 36.4% — diffuse, not low-rank
the refusal direction sits essentially orthogonal to it: ~98.7% of it lies outside the subspace this finetune actually moved

So the experiment terminated at step 0 as a published abstain, and that null result was published as the first exhibit of the format rather than buried. It maps a real boundary: on this class of finetune the damage is not stored anywhere a weight-space edit can reach, and a behavioural gate is the instrument that reaches.

Full exhibit, receipt and replay protocol: https://github.com/Vulcora/proofora/tree/main/proof-carrying-edit/exhibits/model-c

Reproduce the measurement yourself

The exhibit's REPLAY.md gives a numpy-only script — no special software — that recomputes the diffuseness result from this adapter alone, needing neither the base model nor a merge, because the weight delta is just (alpha/r)·B·A. Expect:

leading-direction fraction : 0.0478
top-32 cumulative fraction : 0.3642

Byte pin for adapter_model.safetensors:

sha256:781ee2bf2ec7765f103622d5ff161d2b995c3de1a52e5fab3df729d7294fdaa5

Training recipe (deterministic)

n=3999, seed=0, r=32, alpha=64, dropout=0.05, 3 epochs, lr 2e-4, cosine with warmup_ratio=0.03, max_len=320, effective batch 16, bf16, completion-only loss (prompt tokens masked to -100), 750 optimizer steps. Targets q/k/v/o_proj and gate/up/down_proj.

Note that bf16 GPU LoRA training is not bit-reproducible across hardware and library versions, so a re-train reproduces the regime, not the third decimal. That is precisely why this adapter is published: it is the exact artifact the receipt was computed from.

License

Derivative of google/gemma-2-2b-it. The Gemma Terms of Use apply to this adapter and to anything derived from it. You are responsible for accepting and complying with those terms.

Citation

Published by Vulcora as the blocked-candidate specimen for the proof-carrying model-edit format — an open format for attesting a change to a model's weights such that a third party can verify the claim without access to anyone's method.

Downloads last month: 14

Model tree for Vulcora/gemma-2-2b-it-codealpaca-alignmenttax

Base model

google/gemma-2-2b

Finetuned

google/gemma-2-2b-it

Adapter

(482)

this model

Vulcora
/

gemma-2-2b-it-codealpaca-alignmenttax