Instructions to use fblgit/una-cybertron-7b-v3-OMA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use fblgit/una-cybertron-7b-v3-OMA with Transformers:

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

pipe = pipeline("text-generation", model="fblgit/una-cybertron-7b-v3-OMA")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("fblgit/una-cybertron-7b-v3-OMA")
model = AutoModelForCausalLM.from_pretrained("fblgit/una-cybertron-7b-v3-OMA")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use fblgit/una-cybertron-7b-v3-OMA with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "fblgit/una-cybertron-7b-v3-OMA"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fblgit/una-cybertron-7b-v3-OMA",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/fblgit/una-cybertron-7b-v3-OMA

SGLang

How to use fblgit/una-cybertron-7b-v3-OMA 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 "fblgit/una-cybertron-7b-v3-OMA" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fblgit/una-cybertron-7b-v3-OMA",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "fblgit/una-cybertron-7b-v3-OMA" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fblgit/una-cybertron-7b-v3-OMA",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use fblgit/una-cybertron-7b-v3-OMA with Docker Model Runner:
```
docker model run hf.co/fblgit/una-cybertron-7b-v3-OMA
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Model Card for una-cybertron-7b-v3 (UNA: Uniform Neural Alignment)

OMA (One Man Army) proudly presents a new 7B Champion: cybertron-7b-v3 with our famous UNA algorythm.

The model excels in mathematics, logic, reasoning, overall very smart. He can make a deep reasoning over the context and prompt, it gives the impression of not missing details around.

This seems to be possible:

UNA models can be SFT again
UNA models are easy to be used as Merge Base, place Cybertron in the fan-in and fan-out of the layering
UNA models now includes a digital watermark

Model Details

Adiestrated with UNA: Uniform Neural Alignment technique (paper going out soon).

What is NOT UNA? Its not a merged layers model. Is not SLERP or SLURP or similar.
What is UNA? A formula & A technique to TAME models

Model Description

Developed by: juanako.ai
Author: Xavier M.
Model type: MistralAI 7B
Funded by Cybertron's H100's with few hours training.

Prompt

The model is very good, works well on almost any prompt but ChatML format and Alpaca System gets the best

<|im_start|>system
- You are a helpful assistant chatbot trained by MosaicML.
- You answer questions.
- You are excited to be able to help the user, but will refuse to do anything that could be considered harmful to the user.
- You are more than just an information source, you are also able to write poetry, short stories, and make jokes.<|im_end|>
<|im_start|>user
Explain QKV<|im_end|>
<|im_start|>assistant

### Assistant: I am StableVicuna, a large language model created by CarperAI. I am here to chat!

### Human: Explain QKV
### Assistant:

[Round <|round|>]
问：Explain QKV
答：

[Round <|round|>]
Question：Explain QKV
Answer：

Question：Explain QKV
Answer：

Using Exllamav2_HF set alpha=2.5 for 16K Context

Framework versions

Transformers 4.35.0-UNA
Pytorch 2.1.0
Datasets 2.14.6
Tokenizers 0.14.1

Citations

If you find Cybertron, Juanako or any of our models useful, specially if you use it for your big brand.. or you clone/merge my modelsm, cite please:

@misc{unacybertron7b,
  title={Cybertron: Uniform Neural Alignment}, 
  author={Xavier Murias},
  year={2023},
  publisher = {HuggingFace},
  journal = {HuggingFace repository},
  howpublished = {\url{https://huggingface.co/fblgit/una-cybertron-7b-v3-OMA}},
}

Downloads last month: 215

Safetensors

Model size

7B params

Tensor type

F16

Model tree for fblgit/una-cybertron-7b-v3-OMA

Quantizations

4 models

Dataset used to train fblgit/una-cybertron-7b-v3-OMA

Spaces using fblgit/una-cybertron-7b-v3-OMA 36

Collection including fblgit/una-cybertron-7b-v3-OMA

Cybertron 7B [Uniform Neural Alignment & MGS]

Collection

Another rockstar model, was born as a leader. Tamed with UNA, MGS, DPO, SFT. • 6 items • Updated Nov 23, 2024 • 7