jarodrigues commited on
Commit
e582c02
1 Parent(s): 99677b3

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -6
README.md CHANGED
@@ -31,15 +31,17 @@ datasets:
31
 
32
  # Gervásio 7B PT-PT Instruct
33
 
 
34
 
35
- **Gervásio PT-*** is a ??, large language model for the **Portuguese language**.
36
 
37
- +++++corrigir
38
- It is a **decoder** of the GPT family, based on the neural architecture Transformer and
39
- developed over the Pythia model, with competitive performance for this language.
40
  It has different versions that were trained for different variants of Portuguese (PT),
41
- namely the European variant from Portugal (**PT-PT**) and the American variant from Brazil (**PT-BR**),
42
- and it is distributed free of charge and under a most permissible license.
 
 
43
  ++++++++
44
 
45
  **Gervásio PT-PT 7B Instruct** is developed by NLX-Natural Language and Speech Group, at the University of Lisbon, Faculty of Sciences, Department of Informatics, Portugal.
 
31
 
32
  # Gervásio 7B PT-PT Instruct
33
 
34
+ **Gervásio PT-*** is a competitive **fully open** decoder for the **Portuguese language** language.
35
 
 
36
 
37
+ It is a **decoder** of the GPT family, based on the neural architecture Transformer and developed over the LLaMA~2 7B model.
38
+ Its further improvement through additional training was done over language resources that include new instruction data sets of Portuguese prepared for this purpose.
39
+
40
  It has different versions that were trained for different variants of Portuguese (PT),
41
+ namely the European variant from Portugal (**PT-PT**) and the American variant from Brazil (**PT-BR**).
42
+
43
+ All versions of Gervásio are distributed for free and under a fully open license, including for either research or commercial usage, and can
44
+ be run on consumer-grade hardware, thus seeking to contribute to the advancement of research and innovation in language technology for Portuguese.
45
  ++++++++
46
 
47
  **Gervásio PT-PT 7B Instruct** is developed by NLX-Natural Language and Speech Group, at the University of Lisbon, Faculty of Sciences, Department of Informatics, Portugal.