Update README.md
Browse files
README.md
CHANGED
|
@@ -117,6 +117,33 @@ Here is the standard LLAMA3 template:
|
|
| 117 |
}
|
| 118 |
</PRE>
|
| 119 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 120 |
<b>Optional Enhancement:</B>
|
| 121 |
|
| 122 |
The following can be used in place of the "system prompt" or "system role" to further enhance the model.
|
|
|
|
| 117 |
}
|
| 118 |
</PRE>
|
| 119 |
|
| 120 |
+
<B>Settings: CHAT / ROLEPLAY and/or SMOOTHER operation of this model:</B>
|
| 121 |
+
|
| 122 |
+
In "KoboldCpp" or "oobabooga/text-generation-webui" or "Silly Tavern" ;
|
| 123 |
+
|
| 124 |
+
Set the "Smoothing_factor" to 1.5 to 2.5
|
| 125 |
+
|
| 126 |
+
: in KoboldCpp -> Settings->Samplers->Advanced-> "Smooth_F"
|
| 127 |
+
|
| 128 |
+
: in text-generation-webui -> parameters -> lower right.
|
| 129 |
+
|
| 130 |
+
: In Silly Tavern this is called: "Smoothing"
|
| 131 |
+
|
| 132 |
+
|
| 133 |
+
NOTE: For "text-generation-webui"
|
| 134 |
+
|
| 135 |
+
-> if using GGUFs you need to use "llama_HF" (which involves downloading some config files from the SOURCE version of this model)
|
| 136 |
+
|
| 137 |
+
Source versions (and config files) of my models are here:
|
| 138 |
+
|
| 139 |
+
https://huggingface.co/collections/DavidAU/d-au-source-files-for-gguf-exl2-awq-gptq-hqq-etc-etc-66b55cb8ba25f914cbf210be
|
| 140 |
+
|
| 141 |
+
OTHER OPTIONS:
|
| 142 |
+
|
| 143 |
+
- Increase rep pen to 1.1 to 1.15 (you don't need to do this if you use "smoothing_factor")
|
| 144 |
+
|
| 145 |
+
- If the interface/program you are using to run AI MODELS supports "Quadratic Sampling" ("smoothing") just make the adjustment as noted.
|
| 146 |
+
|
| 147 |
<b>Optional Enhancement:</B>
|
| 148 |
|
| 149 |
The following can be used in place of the "system prompt" or "system role" to further enhance the model.
|