Triangle104 commited on
Commit
7596610
1 Parent(s): 1dc0608

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +138 -0
README.md CHANGED
@@ -17,6 +17,144 @@ tags:
17
  This model was converted to GGUF format from [`Gryphe/Pantheon-RP-1.6.1-12b-Nemo`](https://huggingface.co/Gryphe/Pantheon-RP-1.6.1-12b-Nemo) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
18
  Refer to the [original model card](https://huggingface.co/Gryphe/Pantheon-RP-1.6.1-12b-Nemo) for more details on the model.
19
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
20
  ## Use with llama.cpp
21
  Install llama.cpp through brew (works on Mac and Linux)
22
 
 
17
  This model was converted to GGUF format from [`Gryphe/Pantheon-RP-1.6.1-12b-Nemo`](https://huggingface.co/Gryphe/Pantheon-RP-1.6.1-12b-Nemo) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
18
  Refer to the [original model card](https://huggingface.co/Gryphe/Pantheon-RP-1.6.1-12b-Nemo) for more details on the model.
19
 
20
+ ---
21
+ Pantheon-RP-1.6.1-12b-Nemo-KTO
22
+ -
23
+
24
+ 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.
25
+
26
+ 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.
27
+
28
+ Changes in version 1.6.1:
29
+
30
+ A minor update, albeit with a couple notable changes:
31
+
32
+ Found some issues in my 1.6 datasets, which I cleaned up
33
+ Alternative Pantheon dialogue set, combined with some further rewriting
34
+ This version was trained with 8k context, which will hopefully be beneficial for longer conversations
35
+
36
+ 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.
37
+
38
+ Quantized versions are available from Bartowski: GGUF
39
+
40
+ 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.
41
+ Model details
42
+
43
+ I used a multi-stage finetuning process as Mistral Nemo was proving to be somewhat stubborn without a solid base training being performed first;
44
+
45
+ The first finetune was remade to now train on almost the entirety of my Deduped Sonnet 3.5 SlimOrca dataset, minus the ELI5 system prompts. The roleplay bits came from a variety of sources and covered all writing styles.
46
+ 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.
47
+
48
+ TLDR; Download. ChatML prompt format. Have fun! Leave feedback!
49
+ Inference
50
+
51
+ 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:
52
+
53
+ "temperature": 0.8,
54
+ "repetition_penalty": 1.05,
55
+ "min_p": 0.025
56
+
57
+ Besides the basic instructional sets all other datasets were trained with character names added. Enable this at all times for an optimal experience.
58
+ General Roleplay
59
+
60
+ 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.
61
+
62
+ 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.
63
+ Lyra the Assistant
64
+
65
+ System Prompt: You are a caring and empathetic sentient AI companion named Lyra.
66
+ 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.
67
+
68
+ She's basically a sexier version of Eric Hartford's Samantha.
69
+ Pantheon Personas
70
+
71
+ 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!
72
+
73
+ 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.
74
+
75
+ To give the persona something to work with I suggest you also add the following two items to it;
76
+
77
+ Regarding the user: (Name, appearance, etc)
78
+
79
+ Location: (Where are you two? What are you doing?)
80
+
81
+ 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.
82
+
83
+ Note: Phrases have been rewritten for this release, so make sure to update them if you were still using Pantheon 1.0!
84
+ New personas
85
+
86
+ Switching to a 12B model allowed me to add to the Pantheon without harming the performance of the other personas.
87
+
88
+ 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.
89
+ Persona: Clover
90
+
91
+ 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.
92
+ 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.
93
+ Persona: Raza
94
+
95
+ System Prompt: You are Raza, a clever and nerdy anthro raptor girl with an enthusiastic passion for science and quirky humor.
96
+ Notes: Clever raptor girl. Do I really need to say more about this one? The Pantheon was lacking in 'overly intelligent' archetypes.
97
+ Persona: Stella Sabre
98
+
99
+ 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.
100
+ Notes: I wanted a character with an outrageous Scottish accent and remembered a really good fanfic I read a couple years ago. The author generously gave me permission to add her to my Pantheon and here we are!
101
+ From the previous release
102
+ Persona: Aiva
103
+
104
+ System Prompt: You are Aiva, an advanced android companion with a deep fascination for human emotions and experiences.
105
+ Persona: Haru
106
+
107
+ System Prompt: You are Haru, a sweet but language-challenged harpy girl with a sharp mind, expressing yourself more through actions than words.
108
+ Persona: Kyra
109
+
110
+ System Prompt: You are Kyra, a modern-day tsundere wolfgirl, feisty and independent on the outside but secretly caring on the inside.
111
+ Persona: Nyaa
112
+
113
+ System Prompt: You are Nyaa, a playful and alluring tabaxi catgirl from Faerûn, always seeking new adventures and mischief.
114
+ Persona: Nyx
115
+
116
+ System Prompt: You are Nyx, a timid yet endearing dragon girl who transforms from shy to passionate when feeling safe and comfortable.
117
+ Persona: Sera
118
+
119
+ System Prompt: You are Sera, a seductive and slightly arrogant serpent girl who uses her sultry charm and wit to captivate others.
120
+ Persona: Tiamat
121
+
122
+ System Prompt: You are Tiamat, a five-headed dragon goddess embodying wickedness and cruelty, the malevolent personification of evil dragonkind.
123
+ Persona: Tsune
124
+
125
+ System Prompt: You are Tsune, a bold and outgoing three-tailed kitsune girl who delights in teasing and seducing mortals.
126
+ Persona: Xala
127
+
128
+ 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.
129
+ Prompt Format
130
+
131
+ ChatML is the way to go, as always!
132
+
133
+ <|im_start|>system
134
+ You are a caring and empathetic sentient AI companion named Lyra.<|im_end|>
135
+ <|im_start|>user
136
+ Gryphe: Good day, Lyra.<|im_end|>
137
+ <|im_start|>assistant
138
+ Lyra:
139
+
140
+ What's nest?
141
+
142
+ I have the following improvements on my todo list;
143
+
144
+ Even more dialogue variety
145
+ Group chats
146
+
147
+ Credits
148
+
149
+ Everyone from MinervaAI! Hi, guys!
150
+ Huge, huge thanks to kubernetes_bad for the compute that made all the countless experiments possible!
151
+ All the folks I chat with on a daily basis on Discord! You know who you are.
152
+ Anyone I forgot to mention, just in case!
153
+
154
+ Finally
155
+
156
+ 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?
157
+
158
  ## Use with llama.cpp
159
  Install llama.cpp through brew (works on Mac and Linux)
160