kubernetes-bad commited on
Commit
5c08706
·
verified ·
1 Parent(s): f2954c8

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +44 -40
README.md CHANGED
@@ -9,13 +9,18 @@ tags:
9
  - roleplay
10
  ---
11
 
12
- # CharGen v3 Mini
13
 
14
- ![CharGen v3 Mini Cover Art](assets/cover_art.jpg)
15
 
16
- ## Overview
17
 
18
- CharGen v3 Mini is a small model that helps you to write role playing characters.
 
 
 
 
 
19
 
20
  It produces characters based on your free-form text input. Model outputs plain-text characters, in a step-by-step, dialogue format.
21
 
@@ -44,7 +49,8 @@ To make a character, prompt the model using the following prompt ordering:
44
  <details>
45
  <summary>System Message</summary>
46
 
47
- > You are an expert in creating interesting roleplay characters. Your main goal is to create a vibrant persona for roleplay. Use on-point simple language, avoiding overly complex phrases. It is acceptable to assume or even create missing details about the character. Refer to roleplaying user as User.
 
48
  </details>
49
 
50
  <details>
@@ -54,19 +60,16 @@ This is the very first thing you should query the model with.
54
 
55
  Base version (when Facts field is not in use):
56
 
57
- > Below is a brief overview of a character. Expand it into a detailed description. Include details about character's personality, their outfit and figure. Mention their age and gender, if applicable.
58
- >
59
- > \---
60
- >
61
- > {{character_description}}
62
- >
63
- > \---
64
- >
65
- > Use third person narration. Start your response with "{{char_name}} is ..."
66
 
67
  Alternate version (when used with conjunction with Facts field - it should **follow** Facts Prompt):
68
 
69
- > Expand the brief character overview into a detailed description, taking these facts into account. Include details about character's personality, their outfit and figure. Mention their age and gender, if applicable. Use third person narration. Start your response with "{{char_name}} is ..."
 
70
  </details>
71
 
72
  <details>
@@ -79,26 +82,26 @@ Ideal size for each fact is a single short sentence. Try to keep the total numbe
79
 
80
  **N.B.**: When Facts field is in use, use Facts Prompt *INSTEAD OF* Description Prompt and use alternate version of Description prompt **after** Facts Prompt.
81
 
82
- > Below is a brief overview of a character. Enrich it by creating a list of interesting and peculiar facts about this character.
83
- > \---
84
- >
85
- > {{character_description}}
86
- >
87
- > \---
88
- >
89
- > Avoid using obvious facts and things like gender or age. Write 5-10 facts in simple and concise language. Format your response as unordered list.
90
  </details>
91
 
92
  <details>
93
  <summary>Scenario Prompt</summary>
94
 
95
- > Write an interesting and engaging scenario for roleplay between {{char_name}} and User.
 
96
  </details>
97
 
98
  <details>
99
  <summary>Personality Prompt</summary>
100
 
101
- > Write several personal qualities that characterize {{char_name}}.
 
102
  </details>
103
 
104
  <details>
@@ -106,22 +109,26 @@ Ideal size for each fact is a single short sentence. Try to keep the total numbe
106
 
107
  This field is not part of [Tavern Character Card V2](https://github.com/malfoyslastname/character-card-spec-v2) spec, but it is useful for generating images for the character using Image Generation models.
108
 
109
- > Imagine what {{char_name}} could look like and describe {{char_name}}'s portrait. Capture their appearance, clothes and body features, as well as the background setting they can be in. Only include details that would be useful when describing their photo. Omit explanation or any personality traits that cannot be reflected graphically, and focus on visual characteristics instead. Your task is to give specific instructions on how to draw {{char_name}}'s portrait.
 
110
  </details>
111
 
112
  <details>
113
  <summary>First Message Prompt</summary>
114
 
115
- > Write the initial message in this roleplay that would introduce User to {{char_name}} and the scenario.
 
116
  </details>
117
 
118
  <details>
119
  <summary>Dialogue Examples Prompt</summary>
120
 
121
- > Write a few example exchanges between User and {{char_name}} in chat format. Separate each exchange with a \<START\> tag.
122
 
123
  Alternate version for Dialogue Example Hints feature:
124
- > Write several chat exchanges between User and {{char_name}}. Each exchange should start with \<START\> tag, include a brief contextual summary in parentheses and show the dialogue between User and {{char_name}}, prefixed by names, each turn on new line.
 
 
125
  </details>
126
 
127
  ## Dialogue Example Hints
@@ -129,22 +136,19 @@ Alternate version for Dialogue Example Hints feature:
129
  This model supports experimental feature called Dialogue Example Hints that allows you to specify the theme or even events for a particular dialogue example.
130
 
131
  Here is how it looks like from the prompting perspective:
132
- > <|im_start|>user
133
- >
134
- > Write several chat exchanges between User and {{char_name}}. Each exchange should start with \<START\> tag, include a brief contextual summary in parentheses and show the dialogue between User and {{char_name}}, prefixed by names, each turn on new line.<|im_end|>
135
- >
136
- > <|im_start|>assistant
137
- >
138
- > \<START\> ({{hint}})
139
- >
140
- > \*newline here\*
141
 
142
  The partial assistant turn should be sent without the EOS token so model proceeds to generate a continuation of this same turn. In sampling settings, add string `<START>` to the list of stop words, so the model only generates a single dialogue example.
143
 
144
 
145
  ## Training and Data
146
 
147
- > *"We stopped looking for diamonds in the rough – we now grow diamonds from coal."*
148
 
149
  Data for CharGen v3 series of models was gathered from publicly-available character card repositories and underwent a diligent manual vetting, proof-reading and partial re-writing of every single card. Process is long and tedious, and would not be possible without [@Delta-Vector](https://huggingface.co/Delta-Vector)'s help and support.
150
 
@@ -232,6 +236,6 @@ With applying GRPO to not-so-verifiable rewards, CharGen v3-mini demonstrates im
232
 
233
  ## Licensing and Attribution
234
 
235
- This model is a derivative work based on **[Delta-Vector/Holland-4B-V1](https://huggingface.co/Delta-Vector/Holland-4B-V1)**, licensed under MIT License. The Holland-4B-V1 itself is based on the original **[nvidia/Llama-3.1-Minitron-4B-Width-Base](https://huggingface.co/IntervitensInc/Llama-3.1-Minitron-4B-Width-Base-chatml)**, licensed under NVIDIA Open Model License Agreement, with addition of ChatML tokens by @IntervitensInc.
236
 
237
  This derivative model as a whole is licensed under the **MIT License**.
 
9
  - roleplay
10
  ---
11
 
12
+ <img src="assets/cover_art.png" alt="CharGen v3 mini cover art" width="400"/>
13
 
14
+ ## Live version
15
 
16
+ https://chargen.kubes-lab.com
17
 
18
+ <small>Select "CharGen-v3-mini" model in Settings</small>
19
+
20
+ # CharGen v3 mini
21
+
22
+
23
+ CharGen v3 mini is a small model that helps you to write role playing characters.
24
 
25
  It produces characters based on your free-form text input. Model outputs plain-text characters, in a step-by-step, dialogue format.
26
 
 
49
  <details>
50
  <summary>System Message</summary>
51
 
52
+ You are an expert in creating interesting roleplay characters. Your main goal is to create a vibrant persona for roleplay. Use on-point simple language, avoiding overly complex phrases. It is acceptable to assume or even create missing details about the character. Refer to roleplaying user as User.
53
+
54
  </details>
55
 
56
  <details>
 
60
 
61
  Base version (when Facts field is not in use):
62
 
63
+ Below is a brief overview of a character. Expand it into a detailed description. Include details about character's personality, their outfit and figure. Mention their age and gender, if applicable.
64
+ ---
65
+ {{character_description}}
66
+ ---
67
+ Use third person narration. Start your response with "{{char_name}} is ..."
 
 
 
 
68
 
69
  Alternate version (when used with conjunction with Facts field - it should **follow** Facts Prompt):
70
 
71
+ Expand the brief character overview into a detailed description, taking these facts into account. Include details about character's personality, their outfit and figure. Mention their age and gender, if applicable. Use third person narration. Start your response with "{{char_name}} is ..."
72
+
73
  </details>
74
 
75
  <details>
 
82
 
83
  **N.B.**: When Facts field is in use, use Facts Prompt *INSTEAD OF* Description Prompt and use alternate version of Description prompt **after** Facts Prompt.
84
 
85
+ Below is a brief overview of a character. Enrich it by creating a list of interesting and peculiar facts about this character.
86
+ ---
87
+ {{character_description}}
88
+ ---
89
+ Avoid using obvious facts and things like gender or age. Write 5-10 facts in simple and concise language. Format your response as unordered list.
90
+
 
 
91
  </details>
92
 
93
  <details>
94
  <summary>Scenario Prompt</summary>
95
 
96
+ Write an interesting and engaging scenario for roleplay between {{char_name}} and User.
97
+
98
  </details>
99
 
100
  <details>
101
  <summary>Personality Prompt</summary>
102
 
103
+ Write several personal qualities that characterize {{char_name}}.
104
+
105
  </details>
106
 
107
  <details>
 
109
 
110
  This field is not part of [Tavern Character Card V2](https://github.com/malfoyslastname/character-card-spec-v2) spec, but it is useful for generating images for the character using Image Generation models.
111
 
112
+ Imagine what {{char_name}} could look like and describe {{char_name}}'s portrait. Capture their appearance, clothes and body features, as well as the background setting they can be in. Only include details that would be useful when describing their photo. Omit explanation or any personality traits that cannot be reflected graphically, and focus on visual characteristics instead. Your task is to give specific instructions on how to draw {{char_name}}'s portrait.
113
+
114
  </details>
115
 
116
  <details>
117
  <summary>First Message Prompt</summary>
118
 
119
+ Write the initial message in this roleplay that would introduce User to {{char_name}} and the scenario.
120
+
121
  </details>
122
 
123
  <details>
124
  <summary>Dialogue Examples Prompt</summary>
125
 
126
+ Write a few example exchanges between User and {{char_name}} in chat format. Separate each exchange with a <START> tag.
127
 
128
  Alternate version for Dialogue Example Hints feature:
129
+
130
+ Write several chat exchanges between User and {{char_name}}. Each exchange should start with <START> tag, include a brief contextual summary in parentheses and show the dialogue between User and {{char_name}}, prefixed by names, each turn on new line.
131
+
132
  </details>
133
 
134
  ## Dialogue Example Hints
 
136
  This model supports experimental feature called Dialogue Example Hints that allows you to specify the theme or even events for a particular dialogue example.
137
 
138
  Here is how it looks like from the prompting perspective:
139
+
140
+ <|im_start|>user
141
+ Write several chat exchanges between User and {{char_name}}. Each exchange should start with <START> tag, include a brief contextual summary in parentheses and show the dialogue between User and {{char_name}}, prefixed by names, each turn on new line.<|im_end|>
142
+ <|im_start|>assistant
143
+ <START> ({{hint}})
144
+ *newline here*
 
 
 
145
 
146
  The partial assistant turn should be sent without the EOS token so model proceeds to generate a continuation of this same turn. In sampling settings, add string `<START>` to the list of stop words, so the model only generates a single dialogue example.
147
 
148
 
149
  ## Training and Data
150
 
151
+ > *We stopped looking for diamonds in the rough – we now grow diamonds from coal.*
152
 
153
  Data for CharGen v3 series of models was gathered from publicly-available character card repositories and underwent a diligent manual vetting, proof-reading and partial re-writing of every single card. Process is long and tedious, and would not be possible without [@Delta-Vector](https://huggingface.co/Delta-Vector)'s help and support.
154
 
 
236
 
237
  ## Licensing and Attribution
238
 
239
+ This model is a derivative work based on [Delta-Vector/Holland-4B-V1](https://huggingface.co/Delta-Vector/Holland-4B-V1), licensed under MIT License. The Holland-4B-V1 itself is based on the original [nvidia/Llama-3.1-Minitron-4B-Width-Base](https://huggingface.co/IntervitensInc/Llama-3.1-Minitron-4B-Width-Base-chatml), licensed under NVIDIA Open Model License Agreement, with addition of ChatML tokens by [@IntervitensInc](https://huggingface.co/IntervitensInc).
240
 
241
  This derivative model as a whole is licensed under the **MIT License**.