language:
- gl
- es
- en
- pt
licence:
- MIT
tags:
- Llama
license: llama3.1
base_model:
- meta-llama/Llama-3.1-8B
pipeline_tag: text-generation
library_name: transformers
model-index:
- name: Llama-Carvalho-PT
results:
- task:
type: text-generation
name: Text Generation
dataset:
name: ENEM Challenge (No Images)
type: eduagarcia/enem_challenge
split: train
args:
num_few_shot: 3
metrics:
- type: acc
value: 19.38
name: accuracy
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Nos-PT/Llama-Carvalho-PT
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: BLUEX (No Images)
type: eduagarcia-temp/BLUEX_without_images
split: train
args:
num_few_shot: 3
metrics:
- type: acc
value: 18.92
name: accuracy
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Nos-PT/Llama-Carvalho-PT
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: Assin2 RTE
type: assin2
split: test
args:
num_few_shot: 15
metrics:
- type: f1_macro
value: 87.5
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Nos-PT/Llama-Carvalho-PT
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: Assin2 STS
type: eduagarcia/portuguese_benchmark
split: test
args:
num_few_shot: 15
metrics:
- type: pearson
value: 75.7
name: pearson
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Nos-PT/Llama-Carvalho-PT
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: FaQuAD NLI
type: ruanchaves/faquad-nli
split: test
args:
num_few_shot: 15
metrics:
- type: f1_macro
value: 43.97
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Nos-PT/Llama-Carvalho-PT
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: HateBR Binary
type: ruanchaves/hatebr
split: test
args:
num_few_shot: 25
metrics:
- type: f1_macro
value: 76.93
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Nos-PT/Llama-Carvalho-PT
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: PT Hate Speech Binary
type: hate_speech_portuguese
split: test
args:
num_few_shot: 25
metrics:
- type: f1_macro
value: 49.21
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Nos-PT/Llama-Carvalho-PT
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: tweetSentBR
type: eduagarcia/tweetsentbr_fewshot
split: test
args:
num_few_shot: 25
metrics:
- type: f1_macro
value: 60.88
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=Nos-PT/Llama-Carvalho-PT
name: Open Portuguese LLM Leaderboard
Llama-Carvalho-PT
Table of Contents
Click to expand
Model description
Llama-Carvalho-PT is a 8B-parameter transformer-based causal language model for Galician, Portuguese, Spanish and English. It is the result of a continual pretraining of meta-llama/Llama-3.1-8B with a multilingual corpus of 340M tokens with emphasis in Portuguese.
This model is part of the Carvalho familily, a family of LLMs specialized in Portuguese and Galician which can be found here.
Intended uses and limitations
The Llama-Carvalho-PT model is ready-to-use only for causal language modeling. It can perform text-generation tasks and be fine-tuned for specific scenarios.
How to use
import torch
from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
input_text = "Hoxe fai un bo día. O sol "
model_id = "Nos-PT/Llama-Carvalho-PT"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)
generator = pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
torch_dtype=torch.bfloat16,
trust_remote_code=True,
device_map="auto",
)
generation = generator(
input_text,
do_sample=True,
top_k=10,
eos_token_id=tokenizer.eos_token_id
)
print(f"Result: {generation[0]['generated_text']}")
Training
Tools
It was trained using HuggingFace Transformers and Pytorch, using the Causal Modeling Language script. We also use DeepSpeed to deal with the huge size of the model.
Training data
The training corpus consists of texts in 4 languages, with an emphasis on Portuguese. The main aim of this is to ensure that the model learns to work with this language perfectly, while maintaining knowledge of languages already known (Spanish, English), learning others (Galician) or adapting existing language varieties (Portuguese-PT instead of Portuguese-BR).
The corpus is composed as follows:
| Corpus | gl | pt | es | en | |
|---|---|---|---|---|---|
| Base plain text corpus | Tokens | 30M | 250M | 29M | 29M |
| Percentage (of the total base corpus) | 9% | 74% | 8,5% | 8,5% | |
| Instructions | Tokens | 26,7M | 44M | 804K | 623K |
| Percentage (of the total instructions corpus) | 37,01% | 61,00% | 1,11% | 0,86% |
Training hyperparameters
- seed: 42
- num_devices: 1
- train_batch_size: 4
- eval_batch_size: 4
- gradient_acummulation: 4
- optimizer: AdamW
- betas: (0.9,0.999)
- epsilon: 1e-08
- weight_decay_rate: 0.1
- scheduler: "Linear"
- learning_rate: 1e-04
- num_epochs: 1.0
Framework
The training was conducted on the Vision Clúster in the University of Evora (BSC, using 1 node with 8 GPUs NVIDIA A100 40G.
Evaluation
In process...
Galician and European Portuguese
Soon...
American Portuguese: Open Portuguese LLM Leaderboard Evaluation Results
Detailed results can be found here and on the 🚀 Open Portuguese LLM Leaderboard
| Metric | Value |
|---|---|
| Average | 54.06 |
| ENEM Challenge (No Images) | 19.38 |
| BLUEX (No Images) | 18.92 |
| Assin2 RTE | 87.50 |
| Assin2 STS | 75.70 |
| FaQuAD NLI | 43.97 |
| HateBR Binary | 76.93 |
| PT Hate Speech Binary | 49.21 |
| tweetSentBR | 60.88 |
Additional information
Contact
For further information, please send an email to
License
MIT License
Copyright (c) 2024
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
Funding
This model was development within the Nós Project, funded by the Ministerio para la Transformación Digital y de la Función Pública - Funded by EU – NextGenerationEU within the framework of the project ILENIA with reference 2022/TL22/00215336.