Update README.md
Browse files
README.md
CHANGED
|
@@ -46,8 +46,13 @@ Please note that we tested this model with a 5.0bpw EXL2 quant. Results are not
|
|
| 46 |
## Sampling Settings
|
| 47 |
Here are some settings that work well with this model:
|
| 48 |
```
|
| 49 |
-
|
|
|
|
|
|
|
|
|
|
| 50 |
```
|
|
|
|
|
|
|
| 51 |
|
| 52 |
## Credit
|
| 53 |
Thank you to my team consisting of [@ToastyPigeon](https://huggingface.co/ToastyPigeon), [@Fizzarolli](https://huggingface.co/Fizzarolli), and myself [@elinas](https://huggingface.co/elinas).
|
|
|
|
| 46 |
## Sampling Settings
|
| 47 |
Here are some settings that work well with this model:
|
| 48 |
```
|
| 49 |
+
Temp -> 0.7 - 1.2
|
| 50 |
+
Min P -> 0.025 - 0.05 [temp in order, not last]
|
| 51 |
+
Presence Penalty -> 1.0
|
| 52 |
+
Repetition Penalty range -> 4000
|
| 53 |
```
|
| 54 |
+
Higher temp gives more uniqueness and less repetition. Please do not take these settings as the "best" - your system prompt matters significantly, and if you're roleplaying
|
| 55 |
+
use the Basic system prompt in SillyTavern. You can also try other samplers like Top P.
|
| 56 |
|
| 57 |
## Credit
|
| 58 |
Thank you to my team consisting of [@ToastyPigeon](https://huggingface.co/ToastyPigeon), [@Fizzarolli](https://huggingface.co/Fizzarolli), and myself [@elinas](https://huggingface.co/elinas).
|