metadata
base_model: rhaymison/phi-3-portuguese-tom-cat-4k-instruct
datasets:
- rhaymison/superset
inference: true
language:
- pt
library_name: transformers
license: apache-2.0
model-index:
- name: phi-3-portuguese-tom-cat-4k-instruct
results:
- dataset:
args:
num_few_shot: 3
name: ENEM Challenge (No Images)
split: train
type: eduagarcia/enem_challenge
metrics:
- name: accuracy
type: acc
value: 61.58
source:
name: Open Portuguese LLM Leaderboard
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
task:
name: Text Generation
type: text-generation
- dataset:
args:
num_few_shot: 3
name: BLUEX (No Images)
split: train
type: eduagarcia-temp/BLUEX_without_images
metrics:
- name: accuracy
type: acc
value: 50.63
source:
name: Open Portuguese LLM Leaderboard
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
task:
name: Text Generation
type: text-generation
- dataset:
args:
num_few_shot: 3
name: OAB Exams
split: train
type: eduagarcia/oab_exams
metrics:
- name: accuracy
type: acc
value: 43.69
source:
name: Open Portuguese LLM Leaderboard
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
task:
name: Text Generation
type: text-generation
- dataset:
args:
num_few_shot: 15
name: Assin2 RTE
split: test
type: assin2
metrics:
- name: f1-macro
type: f1_macro
value: 91.54
source:
name: Open Portuguese LLM Leaderboard
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
task:
name: Text Generation
type: text-generation
- dataset:
args:
num_few_shot: 15
name: Assin2 STS
split: test
type: eduagarcia/portuguese_benchmark
metrics:
- name: pearson
type: pearson
value: 75.27
source:
name: Open Portuguese LLM Leaderboard
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
task:
name: Text Generation
type: text-generation
- dataset:
args:
num_few_shot: 15
name: FaQuAD NLI
split: test
type: ruanchaves/faquad-nli
metrics:
- name: f1-macro
type: f1_macro
value: 47.46
source:
name: Open Portuguese LLM Leaderboard
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
task:
name: Text Generation
type: text-generation
- dataset:
args:
num_few_shot: 25
name: HateBR Binary
split: test
type: ruanchaves/hatebr
metrics:
- name: f1-macro
type: f1_macro
value: 83.01
source:
name: Open Portuguese LLM Leaderboard
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
task:
name: Text Generation
type: text-generation
- dataset:
args:
num_few_shot: 25
name: PT Hate Speech Binary
split: test
type: hate_speech_portuguese
metrics:
- name: f1-macro
type: f1_macro
value: 70.19
source:
name: Open Portuguese LLM Leaderboard
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
task:
name: Text Generation
type: text-generation
- dataset:
args:
num_few_shot: 25
name: tweetSentBR
split: test
type: eduagarcia/tweetsentbr_fewshot
metrics:
- name: f1-macro
type: f1_macro
value: 57.78
source:
name: Open Portuguese LLM Leaderboard
url: >-
https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/phi-3-portuguese-tom-cat-4k-instruct
task:
name: Text Generation
type: text-generation
model_creator: rhaymison
model_name: phi-3-portuguese-tom-cat-4k-instruct
pipeline_tag: text-generation
quantized_by: afrideva
tags:
- portugues
- portuguese
- QA
- instruct
- phi
- gguf
- ggml
- quantized
phi-3-portuguese-tom-cat-4k-instruct-GGUF
Quantized GGUF model files for phi-3-portuguese-tom-cat-4k-instruct from rhaymison
Original Model Card:
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.
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]
