sam2ai commited on
Commit
8798ebe
1 Parent(s): 76c88ec

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +62 -0
README.md CHANGED
@@ -1,3 +1,65 @@
1
  ---
2
  license: cc-by-nc-4.0
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: cc-by-nc-4.0
3
+ language:
4
+ - bn
5
  ---
6
+ # Model Card for Model ID
7
+
8
+ [![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/)
9
+
10
+ ## Model description
11
+
12
+ 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.
13
+
14
+ 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.
15
+
16
+
17
+ ## Training hyper-parameters
18
+ | Parameter | Value |
19
+ | ------ | ------ |
20
+ | Batch size | 128 |
21
+ | Learning rate | 3e-4 |
22
+ | Epochs | 5 |
23
+ |Cutoff length | 256 |
24
+ |Weight_decay | 0.001 |
25
+ |Warmup_rate | 0.1 |
26
+ |LR_scheduler | linear |
27
+ |Lora r | 16 |
28
+ |Lora target modules | (q_proj, k_proj, v_proj, o_proj) |
29
+
30
+
31
+ Instructions for running it can be found at https://github.com/OdiaGenAI/GenerativeAI_and_LLM_Odia.
32
+
33
+ ### Licensing Information
34
+
35
+ This work is licensed under a
36
+ [Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License][cc-by-nc-sa].
37
+
38
+ [![CC BY-NC-SA 4.0][cc-by-nc-sa-image]][cc-by-nc-sa]
39
+
40
+ [cc-by-nc-sa]: http://creativecommons.org/licenses/by-nc-sa/4.0/
41
+ [cc-by-nc-sa-image]: https://licensebuttons.net/l/by-nc-sa/4.0/88x31.png
42
+ [cc-by-nc-sa-shield]: https://img.shields.io/badge/License-CC%20BY--NC--SA%204.0-lightgrey.svg
43
+
44
+ ### Citation Information
45
+
46
+ If you find this helpful repository, please consider giving 👏 and citing:
47
+
48
+ ```
49
+ @misc{OdiaGenAI-Bengali-LLM,
50
+ author = {Shantipriya Parida and Sambit Sekhar and Guneet Singh Kohli and Arghyadeep Sen and Shashikanta Sahoo},
51
+ title = {Bengali Instruction-Tuning Model},
52
+ year = {2023},
53
+ publisher = {Hugging Face},
54
+ journal = {Hugging Face repository},
55
+ howpublished = {\url{https://huggingface.co/OdiaGenAI}},
56
+ }
57
+ ```
58
+
59
+ ### Contributions
60
+
61
+ - Shantipriya Parida
62
+ - Sambit Sekhar
63
+ - Guneet Singh Kohli
64
+ - Arghyadeep Sen
65
+ - Shashikanta Sahoo