Update README.md
Browse files
README.md
CHANGED
|
@@ -43,8 +43,9 @@ Addressing the efficay of Quantization and PEFT. Implemented as a personal Proje
|
|
| 43 |
|
| 44 |
```
|
| 45 |
The quantized model is finetuned as PEFT. We have the trained Adapter.
|
| 46 |
-
Merging LoRA
|
| 47 |
-
So instead of loading a single finetuned model, we need to load the
|
|
|
|
| 48 |
```
|
| 49 |
|
| 50 |
```python
|
|
|
|
| 43 |
|
| 44 |
```
|
| 45 |
The quantized model is finetuned as PEFT. We have the trained Adapter.
|
| 46 |
+
Merging LoRA adapater with GPTQ quantized model is not yet supported.
|
| 47 |
+
So instead of loading a single finetuned model, we need to load the base
|
| 48 |
+
model and merge the finetuned adapter on top.
|
| 49 |
```
|
| 50 |
|
| 51 |
```python
|