Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
library_name: peft
|
| 3 |
---
|
|
@@ -17,4 +36,4 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 17 |
### Framework versions
|
| 18 |
|
| 19 |
|
| 20 |
-
- PEFT 0.5.0.dev0
|
|
|
|
| 1 |
+
# Cabra: A portuguese finetuned instruction Open-LLaMA
|
| 2 |
+
|
| 3 |
+
LoRA adapter created with the procedures detailed at the GitHub repository: https://github.com/gustrd/cabra .
|
| 4 |
+
|
| 5 |
+
This training was done at 2 epochs using one A4000 at Paperspace.
|
| 6 |
+
|
| 7 |
+
The GGML version was created with llama.cpp "convert-lora-to-ggml.py".
|
| 8 |
+
|
| 9 |
+
---
|
| 10 |
+
library_name: peft
|
| 11 |
+
license: cc-by-sa-3.0
|
| 12 |
+
datasets:
|
| 13 |
+
- Gustrd/dolly-15k-libretranslate-pt
|
| 14 |
+
language:
|
| 15 |
+
- pt
|
| 16 |
+
---
|
| 17 |
+
|
| 18 |
+
This LoRA adapter was created following the procedure
|
| 19 |
+
|
| 20 |
---
|
| 21 |
library_name: peft
|
| 22 |
---
|
|
|
|
| 36 |
### Framework versions
|
| 37 |
|
| 38 |
|
| 39 |
+
- PEFT 0.5.0.dev0
|