metadata
language:
- pt
license: apache-2.0
library_name: transformers
tags:
- portugues
- portuguese
- QA
- instruct
- phi
base_model: microsoft/Phi-3-mini-4k-instruct
datasets:
- rhaymison/superset
pipeline_tag: text-generation
model-index:
- name: phi-3-portuguese-tom-cat-4k-instruct
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: 61.58
name: accuracy
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
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: 50.63
name: accuracy
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
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: 43.69
name: accuracy
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
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: 91.54
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
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: 75.27
name: pearson
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
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: 47.46
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
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: 83.01
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
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: 70.19
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
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: 57.78
name: f1-macro
source:
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
name: Open Portuguese LLM Leaderboard
Phi-3-portuguese-tom-cat-4k-instruct
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 microsoft/Phi-3-mini-4k.
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:
- rhaymison/phi-3-portuguese-tom-cat-4k-instruct-q8-gguf
- rhaymison/phi-3-portuguese-tom-cat-4k-instruct-f16-gguf
Explore this and other models to find the best fit for your needs!
How to use
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 4b) 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/phi-3-portuguese-tom-cat-4k-instruct", device_map= {"": 0})
tokenizer = AutoTokenizer.from_pretrained("rhaymison/phi-3-portuguese-tom-cat-4k-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_template(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"""<s><|system|>
{ system_prompt }
<|user|>
{ question }
<|assistant|>
"""
question = format_template("E possivel ir de Carro dos Estados unidos ate o japão")
pipe(question)
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}
)
Open Portuguese LLM Leaderboard Evaluation Results
Detailed results can be found here and on the 🚀 Open Portuguese LLM Leaderboard
| Metric | Value |
|---|---|
| Average | 64.57 |
| ENEM Challenge (No Images) | 61.58 |
| BLUEX (No Images) | 50.63 |
| OAB Exams | 43.69 |
| Assin2 RTE | 91.54 |
| Assin2 STS | 75.27 |
| FaQuAD NLI | 47.46 |
| HateBR Binary | 83.01 |
| PT Hate Speech Binary | 70.19 |
| tweetSentBR | 57.78 |
Comments
Any idea, help or report will always be welcome.
email: [email protected]
