Update README.md
Browse files
README.md
CHANGED
|
@@ -15,6 +15,8 @@ It leverages the open-source GPTQModel quantization to achieve 4-bit precision w
|
|
| 15 |
smaller,
|
| 16 |
faster model with minimal performance degradation.
|
| 17 |
|
|
|
|
|
|
|
| 18 |
Ran on a single NVIDIA A100 GPU with 80GB of VRAM.
|
| 19 |
|
| 20 |
*Note* `batch_size` is set quite high as the model is small, you may need to adjust this to your GPU VRAM.
|
|
|
|
| 15 |
smaller,
|
| 16 |
faster model with minimal performance degradation.
|
| 17 |
|
| 18 |
+
NOTE: High perplexity, maybe due to MSE. Non-MSE quant either present or coming.
|
| 19 |
+
|
| 20 |
Ran on a single NVIDIA A100 GPU with 80GB of VRAM.
|
| 21 |
|
| 22 |
*Note* `batch_size` is set quite high as the model is small, you may need to adjust this to your GPU VRAM.
|