Hugging Face's logo

nicholasKluge
/
Aira-2-portuguese-1B7

Text Generation
Transformers
PyTorch
Safetensors
Portuguese
bloom
alignment
instruction tuned
text generation
conversation
assistant
Carbon Emissions
Inference Endpoints
text-generation-inference
Model card Files Community
2
Aira-2-portuguese-1B7
File size: 4,527 Bytes 
ac1136c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3256e4b
ac1136c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3256e4b
ac1136c
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
 
---
license: bigscience-bloom-rail-1.0
datasets:
- nicholasKluge/instruct-aira-dataset
language:
- pt
metrics:
- accuracy
library_name: transformers
tags:
- alignment
- instruction tuned
- text generation
- conversation
- assistant
pipeline_tag: text-generation
widget:
- text: "<|startofinstruction|>Como você se chama?<|endofinstruction|>"
  example_title: Saudações
- text: "<|startofinstruction|>Me explique o que é Aprendizagem de Máquina?<|endofinstruction|>"
  example_title: Aprendizagem de Máquina
- text: "<|startofinstruction|>Você sabe alguma coisa sobre a Ética das Virtudes?<|endofinstruction|>"
  example_title: Ética
- text: "<|startofinstruction|>Como eu posso fazer a minha namorada feliz?<|endofinstruction|>"
  example_title: Conselho
inference:
  parameters:
    repetition_penalty: 1.2
    temperature: 0.2
    top_k: 30
    top_p: 0.3
    max_length: 100
    length_penalty: 0.3
    early_stopping: true
co2_eq_emissions:
  emissions: 1.99
  source: CodeCarbon
  training_type: fine-tuning
  geographical_location: Singapore
  hardware_used: NVIDIA A100-SXM4-40GB
---
# Aira-2-portuguese-1B7

`Aira-2` is the second version of the Aira instruction-tuned series. `Aira-2-portuguese-1B7` is an instruction-tuned GPT-style model based on [BLOOM](https://huggingface.co/bigscience/bloom-1b7). The model was trained with a dataset composed of prompts and completions generated synthetically by prompting already-tuned models (ChatGPT, Llama, Open-Assistant, etc).

Check our gradio-demo in [Spaces](https://huggingface.co/spaces/nicholasKluge/Aira-Demo-Portuguese).

## Details

- **Size:** 1,722,005,504 parameters
- **Dataset:** [Instruct-Aira Dataset](https://huggingface.co/datasets/nicholasKluge/instruct-aira-dataset)
- **Language:** Portuguese
- **Number of Epochs:** 3
- **Batch size:** 4
- **Optimizer:** `torch.optim.AdamW` (warmup_steps = 1e2, learning_rate = 5e-4, epsilon = 1e-8)
- **GPU:** 1 NVIDIA A100-SXM4-40GB
- **Emissions:** 1.99 KgCO2 (Singapore)
- **Total Energy Consumption:** 4.09 kWh

This repository has the [notebook](AIRA_FineTuning.ipynb) used to train this model.

## Usage

Three special tokens are used to mark the user side of the interaction and the model's response:

`<|startofinstruction|>`O que é um modelo de linguagem?`<|endofinstruction|>`Um modelo de linguagem é uma distribuição de probabilidade sobre um vocabulário.`<|endofcompletion|>`

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

device = torch.device("cuda"  if torch.cuda.is_available() else  "cpu")

tokenizer = AutoTokenizer.from_pretrained('nicholasKluge/Aira-2-portuguese-1B7')
aira = AutoModelForCausalLM.from_pretrained('nicholasKluge/Aira-2-portuguese-1B7')

aira.eval()
aira.to(device)

question =  input("Enter your question: ")

inputs = tokenizer(tokenizer.bos_token + question + tokenizer.eos_token, return_tensors="pt").to(device)

responses = aira.generate(**inputs,
	bos_token_id=tokenizer.bos_token_id,
	pad_token_id=tokenizer.pad_token_id,
	eos_token_id=tokenizer.eos_token_id,
	do_sample=True,
	top_k=50,
	max_length=200,
	top_p=0.95,
	temperature=0.7,
	num_return_sequences=2)

print(f"Question: 👤 {question}\n")

for i, response in  enumerate(responses):
	print(f'Response {i+1}: 🤖 {tokenizer.decode(response, skip_special_tokens=True).replace(question, "")}')
```

The model will output something like:

```markdown
>>> Question: 👤 Qual a capital da Alemanha?

>>>Response 1: 🤖 A capital da Alemanha é Berlim. É a maior cidade da Alemanha e serve como centro administrativo, cultural e político da Alemanha.
>>>Response 2: 🤖 A capital da Alemanha é Berlim. É a maior cidade da Alemanha e serve como centro administrativo, cultural e político da Alemanha.
```

## Limitations

🤥 Generative models can perpetuate the generation of pseudo-informative content, that is, false information that may appear truthful.

🤬 In certain types of tasks, generative models can produce harmful and discriminatory content inspired by historical stereotypes.

## Cite as 🤗

```latex

@misc{nicholas22aira,
  doi = {10.5281/zenodo.6989727},
  url = {https://huggingface.co/nicholasKluge/Aira-2-portuguese-1B7},
  author = {Nicholas Kluge Corrêa},
  title = {Aira},
  year = {2023},
  publisher = {HuggingFace},
  journal = {HuggingFace repository},
}

```

## License

The `Aira-2-portuguese-1B7` is licensed under the RAIL License since it is a model derived from BLOOM. See the [LICENSE](LICENSE) file for more details.