README.md

---
language:
- it
license: apache-2.0
tags:
- text-generation-inference
- text generation
---

# Mistral-7B-v0.1 for Italian Language Text Generation

## Model Architecture
The Mistral-7B-v0.1 model is a transformer-based model that can handle a variety of tasks including but not limited to translation, summarization, and text completion. It's particularly designed for the Italian language and can be fine-tuned for specific tasks.

# Evaluation
[Leaderboard Ita LLM](https://huggingface.co/spaces/FinancialSupport/open_ita_llm_leaderboard)

| hellaswag_it acc_norm | arc_it acc_norm | m_mmlu_it 5-shot acc | Average |
|:----------------------| :--------------- | :-------------------- | :------- |
| 0.6734 | 0.5466 | 0.5334 | 0,5844 |

## How to Use
How to utilize my Mistral for Italian text generation

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

MODEL_NAME = "DeepMount00/Mistral-Ita-7b"

model = AutoModelForCausalLM.from_pretrained(MODEL_NAME, torch_dtype=torch.bfloat16).eval()
model.to(device)
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)

def generate_answer(prompt):
    messages = [
        {"role": "user", "content": prompt},
    ]
    model_inputs = tokenizer.apply_chat_template(messages, return_tensors="pt").to(device)
    generated_ids = model.generate(model_inputs, max_new_tokens=200, do_sample=True,
                                          temperature=0.001, eos_token_id=tokenizer.eos_token_id)
    decoded = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
    return decoded[0]

prompt = "Come si apre un file json in python?"
answer = generate_answer(prompt)
print(answer)
```
---
## Developer
[Michele Montebovi]