Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- ko
|
| 4 |
+
- en
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
+
# Updates in EXAONE-3.5
|
| 8 |
+
|
| 9 |
+
## Key Changes
|
| 10 |
+
- **RoPE Scaling Parameter**: Added to support longer `context_length`.
|
| 11 |
+
- **Memory Optimization**: For the 2.4B model, `tie_word_embeddings` is set to `True` for improved memory efficiency.
|
| 12 |
+
|
| 13 |
+
⚠️ Using the original [Llamafy script](https://huggingface.co/maywell/EXAONE-3.0-7.8B-Instruct-Llamafied) as-is may lead to performance degradation.
|
| 14 |
+
|
| 15 |
+
To address this, I have updated the script and uploaded the Llamafied version of the model.
|
| 16 |
+
|
| 17 |
+
## Special Thanks
|
| 18 |
+
|
| 19 |
+
- **[@maywell](https://huggingface.co/maywell)**
|
| 20 |
+
For updating the code and uploading the model.
|
| 21 |
+
|
| 22 |
+
- **LG AI Research**
|
| 23 |
+
For releasing the original model.
|
| 24 |
+
Check out the [original release here](https://huggingface.co/collections/LGAI-EXAONE/exaone-35-674d0e1bb3dcd2ab6f39dbb4).
|