Merge branch 'main' of https://huggingface.co/LDCC/LDCC-SOLAR-10.7B into main
Browse files
README.md
CHANGED
|
@@ -1,3 +1,19 @@
|
|
| 1 |
---
|
| 2 |
license: cc-by-nc-4.0
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: cc-by-nc-4.0
|
| 3 |
+
language:
|
| 4 |
+
- ko
|
| 5 |
---
|
| 6 |
+
|
| 7 |
+
# Model Card for LDCC-SOLAR-10.7B
|
| 8 |
+
|
| 9 |
+
## Developed by : Wonchul Kim ([Lotte Data Communication](https://www.ldcc.co.kr) AI Technical Team)
|
| 10 |
+
|
| 11 |
+
## Hardware and Software
|
| 12 |
+
|
| 13 |
+
* **Hardware**: We utilized an A100x4 * 1 for training our model
|
| 14 |
+
* **Training Factors**: We fine-tuned this model using a combination of the [DeepSpeed library](https://github.com/microsoft/DeepSpeed) and the [HuggingFace TRL Trainer](https://huggingface.co/docs/trl/trainer) / [HuggingFace Accelerate](https://huggingface.co/docs/accelerate/index)
|
| 15 |
+
|
| 16 |
+
## Base Model : [yanolja/KoSOLAR-10.7B-v0.1](https://huggingface.co/yanolja/KoSOLAR-10.7B-v0.1) (This model is no longer supported due to a tokenizer issue.)
|
| 17 |
+
|
| 18 |
+
## Caution
|
| 19 |
+
- If you want to fine-tune this model, it is recommended to use the [tokenizer.json](https://huggingface.co/LDCC/LDCC-SOLAR-10.7B/blob/v1.1/tokenizer.json) and [tokenizer_config.json](https://huggingface.co/LDCC/LDCC-SOLAR-10.7B/blob/v1.1/tokenizer_config.json) files from revision v1.1.
|