Update README.md
Browse files
README.md
CHANGED
|
@@ -26,6 +26,14 @@ Reduction in prompt size, as it understands nuance better.
|
|
| 26 |
|
| 27 |
And as a side effect more context available for output due to reduction in prompt size.
|
| 28 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
Special thanks to the original model creator:
|
| 30 |
|
| 31 |
[ https://huggingface.co/KoboldAI/LLaMA2-13B-Psyfighter2 ]
|
|
|
|
| 26 |
|
| 27 |
And as a side effect more context available for output due to reduction in prompt size.
|
| 28 |
|
| 29 |
+
Note that there will be an outsized difference between quants especially for creative and/or "no right answer" use cases.
|
| 30 |
+
|
| 31 |
+
Because of this it is suggested to download the highest quant you can operate, and it's closest neighbours so to speak.
|
| 32 |
+
|
| 33 |
+
IE: Q4KS, Q4KM, Q5KS as an example.
|
| 34 |
+
|
| 35 |
+
Imatrix Plus versions to be uploaded at a separate repo shortly.
|
| 36 |
+
|
| 37 |
Special thanks to the original model creator:
|
| 38 |
|
| 39 |
[ https://huggingface.co/KoboldAI/LLaMA2-13B-Psyfighter2 ]
|