concedo commited on
Commit
1fc28b8
1 Parent(s): 1b231ba

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +32 -3
README.md CHANGED
@@ -1,3 +1,32 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ language:
4
+ - en
5
+ ---
6
+
7
+ <div align="center">
8
+
9
+ # Mini-Magnum-Unboxed-12B-GGUF
10
+ </div>
11
+
12
+ This is the GGUF quantization of https://huggingface.co/concedo/Mini-Magnum-Unboxed-12B, which was originally finetuned on top of the https://huggingface.co/intervitens/mini-magnum-12b-v1.1 model to correct some minor personal annoyances towards what would otherwise be an excellent model.
13
+
14
+ You can use [KoboldCpp](https://github.com/LostRuins/koboldcpp/releases/latest) to run this model.
15
+
16
+ - **Instruct prompt format changed to Alpaca** - Honestly, I don't know why more models don't use it. If you are an Alpaca format lover like me, this should help.
17
+ - **Instruct Decensoring Applied** - You should not need a jailbreak for a model to obey the user. The model should always do what you tell it to. No need for weird `"Sure, I will"` or kitten-murdering-threat tricks.
18
+ - **Short Conversation Tuning** - For people who like to also be able to *chat* (think chatbot/DM) with a character rather than just roleplay with it. This adds a small dataset of short chat-message conversations.
19
+
20
+ <!-- prompt-template start -->
21
+ ## Prompt template: Alpaca
22
+
23
+ ```
24
+ ### Instruction:
25
+ {prompt}
26
+
27
+ ### Response:
28
+ ```
29
+
30
+ <!-- prompt-template end -->
31
+
32
+ Please leave any feedback or issues that you may have. All credits go to the tuners of the original source mini-magnum-12b-v1.1 model as well as Mistral for the Mistral Nemo base model.