|
--- |
|
language: |
|
- ko |
|
datasets: DopeorNope/OpenOrca-near-dedup-v1 |
|
license: cc-by-nc-sa-4.0 |
|
--- |
|
**(주)미디어그룹사람과숲과 (주)마커의 LLM 연구 컨소시엄에서 개발된 모델입니다** |
|
**The license is `cc-by-nc-sa`.** |
|
|
|
## Model Details |
|
|
|
**Model Developers** SeungyooLee (DopeorNope) |
|
|
|
**Input** Models input text only. |
|
|
|
**Output** Models generate text only. |
|
|
|
**Model Architecture** |
|
pub-llama-13b-v6 is an auto-regressive language model based on the LLaMA2 transformer architecture. |
|
|
|
|
|
## Base Model : [beomi/llama-2-koen-13b](https://huggingface.co/beomi/llama-2-koen-13b) |
|
|
|
**Training Dataset** |
|
DopeorNope/OpenOrca-near-dedup-v1 dataset was created by [Near dedup algorithm](https://arxiv.org/abs/2107.06499) to reduce similarity. |
|
We will open it soon. |
|
|
|
|
|
|
