LeonardPuettmann's picture
Update README.md
6f71931 verified
|
raw
history blame
2.15 kB
metadata
library_name: transformers
base_model:
  - HuggingFaceTB/SmolLM2-1.7B-Instruct
license: apache-2.0
language:
  - en
  - it
tags:
  - translation

SmolMaestra - A tiny Llama model tuned for text translation

   _____                 _ __  __                 _             
  / ____|               | |  \/  |               | |            
 | (___  _ __ ___   ___ | | \  / | __ _  ___  ___| |_ _ __ __ _ 
  \___ \| '_ ` _ \ / _ \| | |\/| |/ _` |/ _ \/ __| __| '__/ _` |
  ____) | | | | | | (_) | | |  | | (_| |  __/\__ \ |_| | | (_| |
 |_____/|_| |_| |_|\___/|_|_|  |_|\__,_|\___||___/\__|_|  \__,_|

Model Card

This model was finetuned with roughly 300.000 examples of translations from English to Italian and Italian to English. The model was finetuned in a way to more directly provide a translation without much explanation.

Finetuning took about 10 hours on an A10G Nvidia GPU.

Due to its size, the model runs very well on CPUs. A very italian Llama model

Usage

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "LeonardPuettmann/SmolMaestra-1.7b-Instruct-v0.1"

model = AutoModelForCausalLM.from_pretrained(
    model_id, 
    device_map="auto",
    trust_remote_code=True,
)

tokenizer = AutoTokenizer.from_pretrained(model_id, add_bos_token=True, trust_remote_code=True)

row_json = [
    {"role": "system", "content": "Your job is to return translations for sentences or words from either Italian to English or English to Italian."},
    {"role": "user", "content": "Do you sell tickets for the bus?"},
]

prompt =  tokenizer.apply_chat_template(row_json, tokenize=False)
model_input = tokenizer(prompt, return_tensors="pt").to("cuda")

with torch.no_grad():
    print(tokenizer.decode(model.generate(**model_input, max_new_tokens=1024)[0]))

Data used

The source for the data were sentence pairs from tatoeba.com. The data can be downloaded from here: https://tatoeba.org/downloads

Credits

Base model: HuggingFaceTB/SmolLM2-1.7B-Instruct Finetuned by: Leonard Püttmann https://www.linkedin.com/in/leonard-p%C3%BCttmann-4648231a9/