language:
- pt
license: apache-2.0
library_name: transformers
tags:
- mixtral
- portuguese
- portugues
base_model: mistralai/Mixtral-8x7B-Instruct-v0.1
datasets:
- rhaymison/superset
pipeline_tag: text-generation
model-index:
- name: Mistral-8x7b-portuguese-luana
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: 69.63
name: accuracy
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-8x7b-portuguese-luana
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: 59.11
name: accuracy
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-8x7b-portuguese-luana
name: Open Portuguese LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: OAB Exams
type: eduagarcia/oab_exams
split: train
args:
num_few_shot: 3
metrics:
- type: acc
value: 49.61
name: accuracy
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-8x7b-portuguese-luana
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: 61.21
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-8x7b-portuguese-luana
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: 79.95
name: pearson
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-8x7b-portuguese-luana
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: 78.6
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-8x7b-portuguese-luana
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: 72.42
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-8x7b-portuguese-luana
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: 73.01
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-8x7b-portuguese-luana
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: 50.9
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-8x7b-portuguese-luana
name: Open Portuguese LLM Leaderboard
Mistral-8x7b-Quantized-portuguese-luanaa
This model was trained with a superset of 300,000 instructions in Portuguese. The model comes to help fill the gap in models in Portuguese. Tuned from the Mistral 8x7b and quantized in 4bit for Portuguese, the model was adjusted mainly for instructional tasks.
How to use
A100 GPU
You can use the model in its normal form up to 4-bit quantization. Below we will use both approaches. Remember that verbs are important in your prompt. Tell your model how to act or behave so that you can guide them along the path of their response. Important points like these help models (even smaller models like 7b) to perform much better.
!pip install -q -U transformers
!pip install -q -U accelerate
!pip install -q -U bitsandbytes
from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
model = AutoModelForCausalLM.from_pretrained("rhaymison/Mistral-8x7b-Quantized-portuguese-luana", device_map= {"": 0})
tokenizer = AutoTokenizer.from_pretrained("rhaymison/Mistral-8x7b-Quantized-portuguese-luana")
model.eval()
You can use with Pipeline but in this example i will use such as Streaming
inputs = tokenizer([f"""<s>[INST] Abaixo está uma instrução que descreve uma tarefa, juntamente com uma entrada que fornece mais contexto.
Escreva uma resposta que complete adequadamente o pedido.
### instrução: aja como um professor de matemática e me explique porque 2 + 2 = 4.
[/INST]"""], return_tensors="pt")
inputs.to(model.device)
streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
_ = model.generate(**inputs, streamer=streamer, max_new_tokens=200)
If you are having a memory problem such as "CUDA Out of memory", you should use 4-bit or 8-bit quantization. For the complete model in colab you will need the A100. If you want to use 4bits or 8bits, T4 or L4 will already solve the problem.
4bits example
from transformers import BitsAndBytesConfig
import torch
nb_4bit_config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_quant_type="nf4",
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True
)
model = AutoModelForCausalLM.from_pretrained(
base_model,
quantization_config=bnb_config,
device_map={"": 0}
)
LangChain
Extracting a langChain conversation with a drug petition document as context.
Human: Bom dia. Me informe quais medicamentos são solicitados no texto. Informe também a posologia e o período de tratamento.
Luana: Olá! No texto são solicitados os medicamentos DULOXETINA 30mg,PREGABALINA 100mg ROSUVASTATINA 10mg. A posologia e o período de tratamento não são especificados, pois isso deve ser indicado periodicamente pelo médico.
Human: Perfeito. E qual é o valor da causa informado no documento?
Luana: O valor da causa informado no documento é R$ 6.132,00 (seis mil cento e três).
Human: Foi mencionado no texto alguma lei?
Luana: Sim, a Lei Complementar nº 80/94 é mencionada no texto. Ela se refere às prerrogativas institucionais da Defensoria Pública, incluindo a intimação pessoal de seus membros, a contagem em dobro dos prazos processuais, a manifestação por cotas nos autos e a dispensa da procuração.
Human: Perfeito! Muito obrigado pela ajuda. Por hora é somente isso =]
Luana: Foi um prazer ajudar! Se você tiver outras dúvidas ou necessitar de mais informações, não hesite em perguntar. Tenha um bom dia!
Open Portuguese LLM Leaderboard Evaluation Results
Detailed results can be found here and on the 🚀 Open Portuguese LLM Leaderboard
| Metric | Value |
|---|---|
| Average | 66.05 |
| ENEM Challenge (No Images) | 69.63 |
| BLUEX (No Images) | 59.11 |
| OAB Exams | 49.61 |
| Assin2 RTE | 61.21 |
| Assin2 STS | 79.95 |
| FaQuAD NLI | 78.60 |
| HateBR Binary | 72.42 |
| PT Hate Speech Binary | 73.01 |
| tweetSentBR | 50.90 |
Comments
Any idea, help or report will always be welcome.
email: rhaymisoncristian@gmail.com
