| --- |
| license: apache-2.0 |
| language: |
| - ru |
| - en |
| - de |
| - es |
| - it |
| - ja |
| - vi |
| - zh |
| - fr |
| - pt |
| - id |
| - ko |
| pipeline_tag: text-generation |
| --- |
| # 🌍 Vulture-180B |
| ***Vulture-180B*** is a further fine-tuned causal Decoder-only LLM built by Virtual Interactive (VILM), on top of the famous **Falcon-180B** by [TII](https://www.tii.ae). We collected a new dataset from news articles and Wikipedia's pages of **12 languages** (Total: **80GB**) and continue the pretraining process of Falcon-180B. Finally, we construct a multilingual instructional dataset following **Alpaca**'s techniques. |
|
|
| While ***Vulture-180B*** is an adapter freely usable under **APACHE-2.0**, **Falcon-180B** itself remains available only under the **[Falcon-180B TII License](https://huggingface.co/spaces/tiiuae/falcon-180b-license/blob/main/LICENSE.txt) and [Acceptable Use Policy](https://huggingface.co/spaces/tiiuae/falcon-180b-license/blob/main/ACCEPTABLE_USE_POLICY.txt)**. Users should ensure any commercial applications based on ***Vulture-180B*** comply with the restrictions on **Falcon-180B**'s use. |
|
|
| *Technical Report coming soon* 🤗 |
|
|
| ## Prompt Format |
|
|
| The reccomended model usage is: |
|
|
| ``` |
| A chat between a curious user and an artificial intelligence assistant. |
| |
| USER:{user's question}<|endoftext|>ASSISTANT: |
| ``` |
|
|
| # Model Details |
| ## Model Description |
| - **Developed by:** [https://www.tii.ae](https://www.tii.ae) |
| - **Finetuned by:** [Virtual Interactive](https://vilm.org) |
| - **Language(s) (NLP):** English, German, Spanish, French, Portugese, Russian, Italian, Vietnamese, Indonesian, Chinese, Japanese and Korean |
| - **Training Time:** 3,000 A100 Hours |
|
|
| ## Acknowledgement |
| - Thanks to **TII** for the amazing **Falcon** as the foundation model. |
| - Big thanks to **Google** for their generous Cloud credits. |
|
|
| ## Out-of-Scope Use |
|
|
| Production use without adequate assessment of risks and mitigation; any use cases which may be considered irresponsible or harmful. |
|
|
| ## Bias, Risks, and Limitations |
|
|
| Vulture-180B is trained on a large-scale corpora representative of the web, it will carry the stereotypes and biases commonly encountered online. |
|
|
| ## Recommendations |
|
|
| We recommend users of Vulture-180B to consider finetuning it for the specific set of tasks of interest, and for guardrails and appropriate precautions to be taken for any production use. |
|
|
| ## How to Get Started with the Model |
|
|
| To run inference with the model in full `bfloat16` precision you need approximately 8xA100 80GB or equivalent. |
|
|
| ```python |
| from transformers import AutoTokenizer, AutoModelForCausalLM |
| import transformers |
| import torch |
| from peft import PeftModel |
| |
| model = "tiiuae/falcon-180b" |
| adapters_name = 'vilm/vulture-180b' |
| |
| tokenizer = AutoTokenizer.from_pretrained(model) |
| m = AutoModelForCausalLM.from_pretrained(model, torch_dtype=torch.bfloat16, device_map="auto" ) |
| m = PeftModel.from_pretrained(m, adapters_name) |
| |
| prompt = "A chat between a curious user and an artificial intelligence assistant.\n\nUSER:Thành phố Hồ Chí Minh nằm ở đâu?<|endoftext|>ASSISTANT:" |
| |
| inputs = tokenizer(prompt, return_tensors="pt").to("cuda") |
| |
| output = m.generate(input_ids=inputs["input_ids"], |
| attention_mask=inputs["attention_mask"], |
| do_sample=True, |
| temperature=0.6, |
| top_p=0.9, |
| max_new_tokens=50,) |
| output = output[0].to("cpu") |
| print(tokenizer.decode(output)) |
| ``` |
