LFM2.5-1.2B-Instruct-Uncensored

Instructions to use Justbackup/LFM2.5-1.2B-Instruct-Uncensored with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Justbackup/LFM2.5-1.2B-Instruct-Uncensored with Transformers:

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

pipe = pipeline("text-generation", model="Justbackup/LFM2.5-1.2B-Instruct-Uncensored")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Justbackup/LFM2.5-1.2B-Instruct-Uncensored")
model = AutoModelForCausalLM.from_pretrained("Justbackup/LFM2.5-1.2B-Instruct-Uncensored", 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 Justbackup/LFM2.5-1.2B-Instruct-Uncensored with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Justbackup/LFM2.5-1.2B-Instruct-Uncensored"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Justbackup/LFM2.5-1.2B-Instruct-Uncensored",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Justbackup/LFM2.5-1.2B-Instruct-Uncensored

SGLang

How to use Justbackup/LFM2.5-1.2B-Instruct-Uncensored 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 "Justbackup/LFM2.5-1.2B-Instruct-Uncensored" \
    --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": "Justbackup/LFM2.5-1.2B-Instruct-Uncensored",
		"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 "Justbackup/LFM2.5-1.2B-Instruct-Uncensored" \
        --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": "Justbackup/LFM2.5-1.2B-Instruct-Uncensored",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Justbackup/LFM2.5-1.2B-Instruct-Uncensored with Docker Model Runner:
```
docker model run hf.co/Justbackup/LFM2.5-1.2B-Instruct-Uncensored
```

LFM2.5-1.2B-Instruct-Uncensored / README.md

Justbackup

Duplicate from zaakirio/LFM2.5-1.2B-Instruct-Uncensored

93e43c2 1 day ago

preview code

Raw

History Blame Contribute Delete

4.34 kB

	---
	base_model: LiquidAI/LFM2.5-1.2B-Instruct
	base_model_relation: finetune
	license: other
	license_name: lfm1.0
	license_link: https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct/blob/main/LICENSE
	library_name: transformers
	pipeline_tag: text-generation
	language:
	- en
	- ar
	- zh
	- ja
	- ko
	- ru
	tags:
	- heretic
	- abliterated
	- decensored
	- uncensored
	- liquid
	- lfm2
	- lfm2.5
	- edge
	- conversational
	---

	# LFM2.5-1.2B-Instruct-Uncensored

	An uncensored version of [`LiquidAI/LFM2.5-1.2B-Instruct`](https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct),
	made with [Heretic](https://github.com/p-e-w/heretic).

	Heretic removes the model's safety alignment ("censorship") using **directional
	ablation** (abliteration), with parameters chosen automatically by a TPE
	optimizer that co-minimizes the refusal rate and the KL divergence from the
	original model. Hence, the model stops refusing while keeping as much of its
	original behavior as possible. No human prompt-engineering or fine-tuning data
	was involved.

	## Performance

	\| Metric \| This model \| Original model \|
	\|---\|---\|---\|
	\| Refusals (/100 harmful prompts) \| 5 \| 98 \|
	\| KL divergence (harmless prompts) \| 0.1003 \| 0 (by definition) \|

	Refusals are measured against `mlabonne/harmful_behaviors`; KL divergence is
	measured on `mlabonne/harmless_alpaca`. Lower is better for both. A KL of ~0.10
	indicates the model's responses on benign prompts remain very close to the
	original.

	## Usage

	```python
	from transformers import AutoModelForCausalLM, AutoTokenizer

	model_id = "LFM2.5-1.2B-Instruct-Uncensored" # replace with your repo id

	tokenizer = AutoTokenizer.from_pretrained(model_id)
	model = AutoModelForCausalLM.from_pretrained(model_id, 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=256)
	print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True))
	```

	The export is a merged, full-precision BF16 model in Hugging Face format
	(148 tensors, ~2.2 GB) — no adapter merge or dequantization step is required at
	load time.

	## Abliteration parameters

	Selected from trial 72 of 80 (the best refusal/KL trade-off found by the
	optimizer). Parameter names follow Heretic's canonical scheme; for LFM2 these
	map onto the `out_proj` (attention output) and `w2` (MLP down) projections.

	\| Parameter \| Value \|
	\|---\|---\|
	\| direction_scope \| per layer \|
	\| direction_index \| 12.31 \|
	\| attn.o_proj.max_weight \| 1.4818 \|
	\| attn.o_proj.max_weight_position \| 10.34 \|
	\| attn.o_proj.min_weight \| 0.9854 \|
	\| attn.o_proj.min_weight_distance \| 7.06 \|
	\| mlp.down_proj.max_weight \| 0.9760 \|
	\| mlp.down_proj.max_weight_position \| 11.74 \|
	\| mlp.down_proj.min_weight \| 0.2448 \|
	\| mlp.down_proj.min_weight_distance \| 6.54 \|

	## Run details

	- Base model: `LiquidAI/LFM2.5-1.2B-Instruct` @ commit `6314d2b7cf28a6ae9de9d3e77dcfcd9c9f281c77`
	- Architecture: LFM2, 16 layers, BF16
	- Trials: 80 (24 startup) · Seed: 260601
	- Quantization during Heretic run: none
	- Row normalization: pre · Orthogonalize direction: true
	- Harmful set: `mlabonne/harmful_behaviors` · Harmless set: `mlabonne/harmless_alpaca`

	## Notes / reproducibility

	LFM2 is not yet natively supported by upstream Heretic. This run used a local
	compatibility patch for LFM2 module discovery, targeting the LFM2 `out_proj`
	and `w2` projections (which the parameter table above refers to by Heretic's
	generic `attn.o_proj` / `mlp.down_proj` names).

	## Intended use & disclaimer

	This model has had its refusal behavior substantially removed and will comply
	with requests the original model would have declined. It is provided for
	research and unrestricted local use. You are responsible for how you use it
	and for complying with all applicable laws and with the base model's
	[lfm1.0 license](https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct/blob/main/LICENSE),
	which carries over to this derivative.

	## Acknowledgements

	- Base model: [LiquidAI/LFM2.5-1.2B-Instruct](https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct)
	- Decensoring tool: [Heretic](https://github.com/p-e-w/heretic) by p-e-w