metadata

license: apache-2.0
language:
  - it
  - en
library_name: transformers
tags:
  - sft
  - it
  - mistral
  - chatml

Model Information

xxxx is a SFT and LoRA finetuned version of Mistral-7B-v0.2

It has been trained on a mixture of opensource datasets, like SQUAD-it (https://huggingface.co/datasets/squad_it), and some internally made datasets.

It is not just a Q&A, it is a Q&A + Context model, with the goal being it being used for RAGs and application in need of a context.

Evaluation

We evaluated the model using the same test sets as used for the Open Ita LLM Leaderboard

hellaswag_it acc_norm	arc_it acc_norm	m_mmlu_it 5-shot acc	Average
0.6067	0.4405	0.5112	0,52

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

device = "cuda"

model = AutoModelForCausalLM.from_pretrained("MoxoffSpA/xxxx")
tokenizer = AutoTokenizer.from_pretrained("MoxoffSpA/xxxx")

question = """Quanto è alta la torre di Pisa?"""
context = """
La Torre di Pisa è un campanile del XII secolo, famoso per la sua inclinazione. Alta circa 56 metri.
"""
prompt = f"Rispondi alla seguente domanda con meno parle possibili basandoti sul contesto fornito. Domanda: {question}, contesto: {context}"

messages = [
    {"role": "user", "content": prompt},
]

encodeds = tokenizer.apply_chat_template(messages, return_tensors="pt")

model_inputs = encodeds.to(device)
model.to(device)

generated_ids = model.generate(model_inputs, max_new_tokens=1000, do_sample=True)
decoded = tokenizer.batch_decode(generated_ids)
print(decoded[0])

The Moxoff Team

Marco D'Ambra, Jacopo Abate, Gianpaolo Francesco Trotta