shantipriya
commited on
Commit
·
a3f3f6b
1
Parent(s):
0701de4
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,70 @@
|
|
1 |
---
|
2 |
library_name: peft
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
library_name: peft
|
3 |
---
|
4 |
+
---
|
5 |
+
license: cc-by-nc-sa-4.0
|
6 |
+
language:
|
7 |
+
- or
|
8 |
+
tags:
|
9 |
+
- code
|
10 |
+
---
|
11 |
+
# Model Card for Model ID
|
12 |
+
|
13 |
+
[![License: CC BY-NC-SA 4.0](https://img.shields.io/badge/License-CC_BY--NC--SA_4.0-lightgrey.svg)](https://creativecommons.org/licenses/by-nc-sa/4.0/)
|
14 |
+
|
15 |
+
## Model description
|
16 |
+
|
17 |
+
odiagenAI-bengali-lora-model-v1 is based on Llama-7b and finetuned with 252k Bengali instruction set. The instruction set is translated data from open-source resources, resulting in good Bengali instruction understanding and response generation capabilities.
|
18 |
+
|
19 |
+
The code of Bengali data generation and other detailed information can be found in our Github project repository: https://github.com/OdiaGenAI/GenerativeAI_and_LLM_Odia.
|
20 |
+
|
21 |
+
|
22 |
+
## Training hyper-parameters
|
23 |
+
| Parameter | Value |
|
24 |
+
| ------ | ------ |
|
25 |
+
| Batch size | 128 |
|
26 |
+
| Learning rate | 3e-4 |
|
27 |
+
| Epochs | 5 |
|
28 |
+
|Cutoff length | 256 |
|
29 |
+
|Weight_decay | 0.001 |
|
30 |
+
|Warmup_rate | 0.1 |
|
31 |
+
|LR_scheduler | linear |
|
32 |
+
|Lora r | 16 |
|
33 |
+
|Lora target modules | (q_proj, k_proj, v_proj, o_proj) |
|
34 |
+
|
35 |
+
|
36 |
+
Instructions for running it can be found at https://github.com/OdiaGenAI/GenerativeAI_and_LLM_Odia.
|
37 |
+
|
38 |
+
### Licensing Information
|
39 |
+
|
40 |
+
This work is licensed under a
|
41 |
+
[Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License][cc-by-nc-sa].
|
42 |
+
|
43 |
+
[![CC BY-NC-SA 4.0][cc-by-nc-sa-image]][cc-by-nc-sa]
|
44 |
+
|
45 |
+
[cc-by-nc-sa]: http://creativecommons.org/licenses/by-nc-sa/4.0/
|
46 |
+
[cc-by-nc-sa-image]: https://licensebuttons.net/l/by-nc-sa/4.0/88x31.png
|
47 |
+
[cc-by-nc-sa-shield]: https://img.shields.io/badge/License-CC%20BY--NC--SA%204.0-lightgrey.svg
|
48 |
+
|
49 |
+
### Citation Information
|
50 |
+
|
51 |
+
If you find this helpful repository, please consider giving 👏 and citing:
|
52 |
+
|
53 |
+
```
|
54 |
+
@misc{OdiaGenAI,
|
55 |
+
author = {Shantipriya Parida and Sambit Sekhar and Guneet Singh Kohli and Arghyadeep Sen and Shashikanta Sahoo},
|
56 |
+
title = {Bengali Instruction-Tuning Model},
|
57 |
+
year = {2023},
|
58 |
+
publisher = {Hugging Face},
|
59 |
+
journal = {Hugging Face repository},
|
60 |
+
howpublished = {\url{https://huggingface.co/OdiaGenAI}},
|
61 |
+
}
|
62 |
+
```
|
63 |
+
|
64 |
+
### Contributions
|
65 |
+
|
66 |
+
- Shantipriya Parida
|
67 |
+
- Sambit Sekhar
|
68 |
+
- Guneet Singh Kohli
|
69 |
+
- Arghyadeep Sen
|
70 |
+
- Shashikanta Sahoo
|