README.md

---
tags:
- autotrain
- text-generation
license: openrail
language:
- pl
- en
pipeline_tag: text-generation
---

# Pigeon-7B -- Polish Llama 2

The new Pigeon-7B model is a finetuned Llama 2, trained on over 70k conversational Polish samples.
This is the repository for the 7B fine-tuned model, optimized for question answering and instruction executing.

# Example use:
```
from transformers import LlamaTokenizer
import transformers
import torch

model="Typly/Pigeon-7B"

tokenizer = LlamaTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

prompt=f"""
### Instruction:
Odpisz na wiadomość:
### Input:
Jaka jest róznica pomiędzy playstation i komputerem stacjonarnym?
### Response:
"""

sequences = pipeline(
    prompt,
    do_sample=True,
    temperature=0.1,
    num_return_sequences=1,
    eos_token_id=tokenizer.eos_token_id,
    max_length=len(tokenizer(prompt)['input_ids']) + 100,
)
for seq in sequences:
        print(f"{seq['generated_text'].replace(prompt, '').split('###')[0].strip()}")
```

# Ethical Considerations and Limitations
Pigeon, same as a Llama 2, is a new technology that carries risks with use. Testing conducted to date has been in Polish and English, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, Pigeon’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Pigeon, developers should perform safety testing and tuning tailored to their specific applications of the model.

Please see the Meta's Responsible Use Guide available at [https://ai.meta.com/llama/responsible-use-guide/](https://ai.meta.com/llama/responsible-use-guide)

# Authors

The model was trained by NLP Research Team at Typly.

You can contact us [here](https://typly.app/contact/).