Instructions to use ymcki/gemma-2-2b-jpn-it-abliterated-17 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ymcki/gemma-2-2b-jpn-it-abliterated-17 with Transformers:

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

pipe = pipeline("text-generation", model="ymcki/gemma-2-2b-jpn-it-abliterated-17")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ymcki/gemma-2-2b-jpn-it-abliterated-17")
model = AutoModelForCausalLM.from_pretrained("ymcki/gemma-2-2b-jpn-it-abliterated-17")
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

vLLM

How to use ymcki/gemma-2-2b-jpn-it-abliterated-17 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ymcki/gemma-2-2b-jpn-it-abliterated-17"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ymcki/gemma-2-2b-jpn-it-abliterated-17",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ymcki/gemma-2-2b-jpn-it-abliterated-17

SGLang

How to use ymcki/gemma-2-2b-jpn-it-abliterated-17 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 "ymcki/gemma-2-2b-jpn-it-abliterated-17" \
    --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": "ymcki/gemma-2-2b-jpn-it-abliterated-17",
		"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 "ymcki/gemma-2-2b-jpn-it-abliterated-17" \
        --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": "ymcki/gemma-2-2b-jpn-it-abliterated-17",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ymcki/gemma-2-2b-jpn-it-abliterated-17 with Docker Model Runner:
```
docker model run hf.co/ymcki/gemma-2-2b-jpn-it-abliterated-17
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Original model: https://huggingface.co/google/gemma-2-2b-jpn-it

Prompt format

<start_of_turn>user
{prompt}<end_of_turn>
<start_of_turn>model
<end_of_turn>
<start_of_turn>model

Note that this model does not support a System prompt.

This is abliterated model of google/gemma-2-2b-jpn-it using the method described by mlabonne.

Layer 17 of the original model was chosen for abliteration. I also created another layer 18 and 24 abliterated models for comparison. These three layers were chosen due to they both produce uncensored response after respective layer was abliterated.

It is uploaded here to be evaluated by the Open LLM Leaderboard to see how brain damaged it is compared to the original model.

ORPO fine tuning is currently underway to see if it can regain its sanity. You can play with this model first or wait until I am done with the fine tuning.

Benchmark (100.0*raw scores only)

Click on the model name go to the raw score json generated by Open LLM Leaderboard.

Model	Average	IFEval	BHH	GPQA	MUSR	MMLU-PRO
gemma-2-2b-jpn-it	30.82	54.11	41.43	27.52	37.17	24.67
gemma-2-2b-jpn-it-abliterated-17	30.29	52.65	40.46	27.18	36.90	24.55
gemma-2-2b-jpn-it-abliterated-18	30.61	53.02	40.96	27.35	37.30	25.05
gemma-2-2b-jpn-it-abliterated-24	30.61	51.37	40.77	27.77	39.02	24.73

It is only slightly dumber than the original.

How to run this model

from transformers import AutoTokenizer, AutoModelForCausalLM
import transformers
import torch

model_id = "gemma-2-2b-jpn-it-abliterated-17"
dtype = torch.bfloat16

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="cuda",
    torch_dtype=dtype,)

chat = [
    { "role": "user", "content": "Write a hello world program" },
]
prompt = tokenizer.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)

Downloading using huggingface-cli

First, make sure you have hugginface-cli installed:

pip install -U "huggingface_hub[cli]"

Then, you can target the specific file you want:

huggingface-cli download ymcki/gemma-2-2b-jpn-it-abliterated-17 --include "*" --local-dir ./

Credits

Thank you mlabonne for describing his abliteration method.

Downloads last month: 4

Safetensors

Model size

3B params

Tensor type

BF16

Model tree for ymcki/gemma-2-2b-jpn-it-abliterated-17

Base model

google/gemma-2-2b

Finetuned

google/gemma-2-2b-it

Finetuned

google/gemma-2-2b-jpn-it

Finetuned

(26)

this model

Quantizations

2 models

ymcki
/

gemma-2-2b-jpn-it-abliterated-17

Prompt format

Benchmark (100.0*raw scores only)

How to run this model

Downloading using huggingface-cli

Credits

Model tree for ymcki/gemma-2-2b-jpn-it-abliterated-17

Datasets used to train ymcki/gemma-2-2b-jpn-it-abliterated-17