Commit
•
2d6c63f
1
Parent(s):
c7976c1
Update README.md
Browse files
README.md
CHANGED
|
@@ -13,6 +13,8 @@ I'm interested in companies based outside of South Korea that can provide resour
|
|
| 13 |
Please note that companies with Korean employees, offices in South Korea, or a specific focus on AI-related businesses in South Korea, as well as those engaged in hiring within South Korea, are not eligible. These exclusions will be clearly outlined and confirmed through the drafting and signing of a legally binding contract.
|
| 14 |
|
| 15 |
# The project is currently in progress. Please refrain from using weights and datasets.
|
|
|
|
|
|
|
| 16 |
|
| 17 |
## Status: 19.7k check point weights open, waiting for the results on the LLM leaderboard.
|
| 18 |
|
|
|
|
| 13 |
Please note that companies with Korean employees, offices in South Korea, or a specific focus on AI-related businesses in South Korea, as well as those engaged in hiring within South Korea, are not eligible. These exclusions will be clearly outlined and confirmed through the drafting and signing of a legally binding contract.
|
| 14 |
|
| 15 |
# The project is currently in progress. Please refrain from using weights and datasets.
|
| 16 |
+
- We are currently conducting experiments using various techniques such as max sequence length, rope scaling, attention sinks, and flash attention 2. Please do not use the current model weights as they are not useful. In respect of the licensing of the dataset used for training, we have adopted a non-commercial use (CC-BY-NC-4.0) license. Once the training is complete, we will provide information about the datasets used along with the official release.
|
| 17 |
+
- For `GORANI`, it is intended for research purposes, and for the Korean language model, `KORANI`, it can be used under a commercial use license.
|
| 18 |
|
| 19 |
## Status: 19.7k check point weights open, waiting for the results on the LLM leaderboard.
|
| 20 |
|