rhaymison/Llama-portuguese-13b-Luana-v0.2

Mistral-portuguese-luana-7b

This model was trained with a superset of 200,000 instructions in Portuguese. The model comes to help fill the gap in models in Portuguese. Tuned from the Llama 2 13b in Portuguese, the model was adjusted mainly for instructional tasks. The model comes from the idea of helping to fill the need for Portuguese language models.

How to use

You can use the model in its normal form up to 4-bit or 8-bit quantization. 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 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
)

from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
model = AutoModelForCausalLM.from_pretrained("rhaymison/Llama-portuguese-13b-Luana-v0.2", quantization_config=bnb_config, device_map= {"": 0})
tokenizer = AutoTokenizer.from_pretrained("rhaymison/Llama-portuguese-13b-Luana-v0.2")
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")

streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
_ = model.generate(**inputs, streamer=streamer, max_new_tokens=200)

Open Portuguese LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Average	48.83
ENEM Challenge (No Images)	36.95
BLUEX (No Images)	32.68
OAB Exams	33.30
Assin2 RTE	65.83
Assin2 STS	42.81
FaQuAD NLI	40.44
HateBR Binary	83.62
PT Hate Speech Binary	54.62
tweetSentBR	49.25

Comments

Any idea, help or report will always be welcome.

email: [email protected]