🚀 Llama-3-portuguese-Tom-cat-8b-instruct
This model is fine - tuned from Llama3 8B, trained with a Portuguese chat superset to fill the gap in Portuguese models and is mainly adjusted for chat.
🚀 Quick Start
If you are looking for enhanced compatibility, the Luana model also has a GGUF family that can be run with LlamaCpp.
You can explore the GGUF models starting with the one below:
Explore this and other models to find the best fit for your needs!
This model was trained with a superset of 300,000 chats in Portuguese.
The model comes to help fill the gap in models in Portuguese. Tuned from the Llama3 8B, the model was adjusted mainly for chat.
💻 Usage Examples
Basic Usage
!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/Llama-3-portuguese-Tom-cat-8b-instruct", device_map= {"": 0})
tokenizer = AutoTokenizer.from_pretrained("rhaymison/Llama-3-portuguese-Tom-cat-8b-instruct")
model.eval()
You can use with Pipeline.
from transformers import pipeline
pipe = pipeline("text-generation",
model=model,
tokenizer=tokenizer,
do_sample=True,
max_new_tokens=512,
num_beams=2,
temperature=0.3,
top_k=50,
top_p=0.95,
early_stopping=True,
pad_token_id=tokenizer.eos_token_id,
)
def format_prompt(question:str):
system_prompt = "Abaixo está uma instrução que descreve uma tarefa, juntamente com uma entrada que fornece mais contexto. Escreva uma resposta que complete adequadamente o pedido."
return f"""<|begin_of_text|><|start_header_id|>system<|end_header_id|>
{ system_prompt }<|eot_id|><|start_header_id|>user<|end_header_id|>
{ question }<|eot_id|><|start_header_id|>assistant<|end_header_id|>"""
prompt = format_prompt("Me fale sobra a OAB, Ordem dos Advogados do Brasil")
result = pipe(prompt)
result[0]["generated_text"].split("assistant<|end_header_id|>")[1]
prompt = format_prompt("Haja como um professor de matemática e me explique porque 2 + 2 = 4")
result = pipe(prompt)
result[0]["generated_text"].split("assistant<|end_header_id|>")[1]
Advanced Usage
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}
)
📚 Documentation
Hardware Requirements
- FULL MODEL : A100
- HALF MODEL: L4
- 8bit or 4bit : T4 or V100
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 8b) to perform much better.
📊 Open Portuguese LLM Leaderboard Evaluation Results
Detailed results can be found here and on the 🚀 Open Portuguese LLM Leaderboard
Metric |
Value |
Average |
70.57 |
ENEM Challenge (No Images) |
70.40 |
BLUEX (No Images) |
58 |
OAB Exams |
51.07 |
Assin2 RTE |
90.91 |
Assin2 STS |
75.40 |
FaQuAD NLI |
76.05 |
HateBR Binary |
86.99 |
PT Hate Speech Binary |
60.39 |
tweetSentBR |
65.92 |
Comments
Any idea, help or report will always be welcome.
email: rhaymisoncristian@gmail.com
📄 License
This model is licensed under the Apache-2.0 license.