Commit
•
cc50d3a
1
Parent(s):
35cec9c
Update README.md
Browse files
README.md
CHANGED
@@ -95,6 +95,15 @@ On _free version_ of Google Colab, you may face RAM problems. I guess using `low
|
|
95 |
|
96 |
## Known Issues
|
97 |
|
|
|
|
|
|
|
|
|
|
|
|
|
98 |
## Our Team
|
99 |
|
|
|
|
|
|
|
100 |
## Special Thanks
|
|
|
95 |
|
96 |
## Known Issues
|
97 |
|
98 |
+
* The model produces GPT-3.5 level answers in terms of grammar (specially Persian) but is capable of extremely insane hallucinations. This problem can be solved by a better dataset and better training procedures (such as DPO).
|
99 |
+
* According to the previous issue, the model can also generate misinforming answers specially when dealing with _reasoning_ problems in Persian.
|
100 |
+
* The model is huge, so it requires a lot of resources in order to work correctly. However, we may provide _GPTQ_ or _GGUF_ versions as well.
|
101 |
+
* The prompt format works and it proves our concept of a _instruct following_ LLM, but since we haven't changed `eos_token` and `bos_token` to our own, you may see unncessary information being generated by the model.
|
102 |
+
* According to the previous issue, the model is capable of repeating itself. To solve this problem _temporarily_ you have to keep temperature below 1. According to our tests somewhere between 0.5 to 0.7 is a sweet spot.
|
103 |
+
|
104 |
## Our Team
|
105 |
|
106 |
+
* Muhammadreza Haghiri ([Website](https://haghiri75.com/en) - [Github](https://github.com/prp-e) - [LinkedIn](https://www.linkedin.com/in/muhammadreza-haghiri-1761325b))
|
107 |
+
* Mahi Mohrechi ([Website](https://mohrechi-portfolio.vercel.app/) - [Github](https://github.com/f-mohrechi) - [LinkedIn](https://www.linkedin.com/in/faeze-mohrechi/))
|
108 |
+
|
109 |
## Special Thanks
|