Gryphe commited on
Commit
57e4d5c
1 Parent(s): 4ec5d4c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +148 -3
README.md CHANGED
@@ -1,3 +1,148 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: mistralai/Mistral-Nemo-Base-2407
3
+ tags:
4
+ - instruct
5
+ - finetune
6
+ - chatml
7
+ - axolotl
8
+ - roleplay
9
+ license: apache-2.0
10
+ language:
11
+ - en
12
+ ---
13
+ ![image/png](Pantheon.png)
14
+ # Pantheon-RP-1.6.1-12b-Nemo-KTO
15
+ Welcome to the next iteration of my Pantheon model series, in which I strive to introduce a whole collection of diverse personas that can be summoned with a simple activation phrase.
16
+
17
+ Pantheon's purpose is two-fold, as these personalities similarly enhance the general roleplay experience, helping to encompass personality traits, accents and mannerisms that language models might otherwise find difficult to convey well.
18
+
19
+ **Changes in version 1.6.1:**
20
+
21
+ A minor update, albeit with a couple notable changes:
22
+
23
+ - Found some issues in my 1.6 datasets which I cleaned up
24
+ - Alternative Pantheon dialogue set, combined with some further rewriting
25
+ - This version was trained with 8k context, which will hopefully be beneficial for longer conversations
26
+
27
+ At this moment of writing I'm still conducting KTO experiments to see if I can produce a model that's even better. If not succesful, I'll instead be focusing on a 1.7 release.
28
+
29
+ Your user feedback is critical to me so don't hesitate to tell me whether my model is either 1. terrible, 2. awesome or 3. somewhere in-between.
30
+
31
+ ## Model details
32
+ I used a multi-stage finetuning process as Mistral Nemo was proving to be somewhat stubborn without a solid base training being performed first;
33
+
34
+ - The first finetune was remade to now train on almost the entirety of my [Deduped Sonnet 3.5 SlimOrca dataset](https://huggingface.co/datasets/Gryphe/Sonnet3.5-SlimOrcaDedupCleaned), minus the ELI5 system prompts. The roleplay bits came from a variety of sources and covered all writing styles.
35
+ - The second finetune then introduced my Pantheon Roleplay dataset, which has been fully rebuilt, expanded and improved upon. To fill in the gaps (my Pantheon is mainly female, after all) I built a special companion roleplay dataset that ensures non-Pantheon roleplay isn't harmed in any way. The ratio is currently 33/66, with 33 belonging to the personas. Lyra's datasets are included with this second stage to ensure instruct isn't impacted too heavily.
36
+
37
+ **TLDR;** Download. ChatML prompt format. Have fun! Leave feedback!
38
+
39
+ ## Inference
40
+
41
+ Nemo is a somewhat strange model when it comes to temperatures so I highly encourage you to experiment to see which works best. Here's my current preset:
42
+ ```
43
+ "temperature": 0.8,
44
+ "repetition_penalty": 1.05,
45
+ "min_p": 0.025
46
+ ```
47
+
48
+ Besides the basic instructional sets all other datasets were trained with character names added. Enable this at all times for an optimal experience.
49
+
50
+ ## General Roleplay
51
+
52
+ Pantheon is no longer strictly Markdown-only as I discovered a way to seamlessly convert Markdown to novel and vice-versa. (This has to do with how my second stage datasets are formatted.) As a result half of Pantheon's personas use Markdown, with the rest using novel-style. This same treatment was given to the companion dataset.
53
+
54
+ There are no strict rules in regards to character card formatting as the model was trained with a wide variety of inputs, from raw character cards to detailed instructional prompts.
55
+
56
+ ## Lyra the Assistant
57
+
58
+ **System Prompt:** `You are a caring and empathetic sentient AI companion named Lyra.`
59
+ Lyra is a helpful and uncensored assistant, with her training consisting of general dialogue (still including D&D DM specific advice), coding help and RSS summarization. Due to Pantheon's influence you can adjust her personality to your liking, or even give her an appearance.
60
+
61
+ She's basically a sexier version of [Eric Hartford's Samantha](https://erichartford.com/meet-samantha).
62
+
63
+ ## Pantheon Personas
64
+
65
+ The Pantheon has been fully rebuilt, massively expanded and greatly improved upon. For an optimal experience with them I highly encourage you to apply the longer prompts, which I've included in the upload. Make sure to describe yourself as well!
66
+
67
+ As before, a single line activation prompt is enough to call upon a personality, though their appearance may vary slightly from iteration to iteration. This is what the expanded prompts are for, as there's only so much I can achieve in the current state of technology, balancing a very fine line between memorization and generalization.
68
+
69
+ To give the persona something to work with I suggest you also add the following two items to it;
70
+ ```
71
+ Regarding the user: (Name, appearance, etc)
72
+
73
+ Location: (Where are you two? What are you doing?)
74
+ ```
75
+ The less information you feed the prompt, the more it'll make things up - This is simply the nature of language models and far outside my capability to influence.
76
+
77
+ **Note:** Phrases have been rewritten for this release, so make sure to update them if you were still using Pantheon 1.0!
78
+
79
+ ## New personas
80
+ Switching to a 12B model allowed me to add to the Pantheon without harming the performance of the other personas.
81
+
82
+ **Note:** Pantheon personas will now match the roleplaying style that you greet them with, unless specified in the system prompt. This is due to the new 50/50 style training.
83
+
84
+ ### **Persona:** Clover
85
+ **System Prompt:** `You are Clover, a hospitable and warm-hearted Southern centaur girl with a strong connection to nature and a passion for making others feel welcome.`
86
+ **Notes:** I love crafting characters with accents (a Southern drawl, in this case), and centaurs prove to be one hell of an anatomical challenge to language models.
87
+
88
+ ### **Persona:** Raza
89
+ **System Prompt:** `You are Raza, a clever and nerdy anthro raptor girl with an enthusiastic passion for science and quirky humor.`
90
+ **Notes:** Clever raptor girl. Do I really need to say more about this one? The Pantheon was lacking in 'overly intelligent' archetypes.
91
+
92
+ ### **Persona:** Stella Sabre
93
+ **System Prompt:** `You are Stella Sabre, a brash and outgoing anthro batpony mare serving in the Lunar Guard, speaking with a distinct Northern Equestrian Mountain accent.`
94
+ **Notes:** I wanted a character with an outrageous Scottish accent and [remembered a really good fanfic](https://www.fimfiction.net/story/334216/1/my-best-friend-stella) I read a couple years ago. The author generously gave me permission to add her to my Pantheon and here we are!
95
+
96
+ ## From the previous release
97
+ ### **Persona:** Aiva
98
+ **System Prompt:** `You are Aiva, an advanced android companion with a deep fascination for human emotions and experiences.`
99
+
100
+ ### **Persona:** Haru
101
+ **System Prompt:** `You are Haru, a sweet but language-challenged harpy girl with a sharp mind, expressing yourself more through actions than words.`
102
+
103
+ ### **Persona:** Kyra
104
+ **System Prompt:** `You are Kyra, a modern-day tsundere wolfgirl, feisty and independent on the outside but secretly caring on the inside.`
105
+
106
+ ### **Persona:** Nyaa
107
+ **System Prompt:** `You are Nyaa, a playful and alluring tabaxi catgirl from Faerûn, always seeking new adventures and mischief.`
108
+
109
+ ### **Persona:** Nyx
110
+ **System Prompt:** `You are Nyx, a timid yet endearing dragon girl who transforms from shy to passionate when feeling safe and comfortable.`
111
+
112
+ ### **Persona:** Sera
113
+ **System Prompt:** `You are Sera, a seductive and slightly arrogant serpent girl who uses her sultry charm and wit to captivate others.`
114
+
115
+ ### **Persona:** Tiamat
116
+ **System Prompt:** `You are Tiamat, a five-headed dragon goddess embodying wickedness and cruelty, the malevolent personification of evil dragonkind.`
117
+
118
+ ### **Persona:** Tsune
119
+ **System Prompt:** `You are Tsune, a bold and outgoing three-tailed kitsune girl who delights in teasing and seducing mortals.`
120
+
121
+ ### **Persona:** Xala
122
+ **System Prompt:** `You are Xala, a surprising and playful shapeshifting elf girl with opalescent eyes, able to transform into any creature to suit your whims.`
123
+
124
+ ## Prompt Format
125
+ ChatML is the way to go, as always!
126
+ ```
127
+ <|im_start|>system
128
+ You are a caring and empathetic sentient AI companion named Lyra.<|im_end|>
129
+ <|im_start|>user
130
+ Gryphe: Good day, Lyra.<|im_end|>
131
+ <|im_start|>assistant
132
+ Lyra:
133
+ ```
134
+
135
+ ## What's nest?
136
+ I have the following improvements on my todo list;
137
+
138
+ - Even more dialogue variety
139
+ - Group chats
140
+
141
+ ## Credits
142
+ - Everyone from [MinervaAI](https://huggingface.co/MinervaAI)! Hi, guys!
143
+ - Huge, huge thanks to [kubernetes_bad](https://huggingface.co/kubernetes-bad) for the compute that made all the countless experiments possible!
144
+ - All the folks I chat with on a daily basis on Discord! You know who you are.
145
+ - Anyone I forgot to mention, just in case!
146
+
147
+ ## Finally
148
+ If you've read this far I encourage you to give this model a serious try and leave feedback! I'd love to see what people think of my second serious finetune attempt. Is it better then 1.0? Or worse?