TheBloke commited on
Commit
0afc457
1 Parent(s): a5a9f9b

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +19 -16
README.md CHANGED
@@ -23,7 +23,7 @@ pipeline_tag: text-generation
23
  </div>
24
  <div style="display: flex; justify-content: space-between; width: 100%;">
25
  <div style="display: flex; flex-direction: column; align-items: flex-start;">
26
- <p><a href="https://discord.gg/UBgz4VXf">Chat & support: my new Discord server</a></p>
27
  </div>
28
  <div style="display: flex; flex-direction: column; align-items: flex-end;">
29
  <p><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
@@ -100,7 +100,9 @@ Further instructions here: [text-generation-webui/docs/llama.cpp-models.md](http
100
  <!-- footer start -->
101
  ## Discord
102
 
103
- For further support, and discussions on these models and AI in general, join us at: [TheBloke AI's Discord server](https://discord.gg/UBgz4VXf)
 
 
104
 
105
  ## Thanks, and how to contribute.
106
 
@@ -108,15 +110,16 @@ Thanks to the [chirper.ai](https://chirper.ai) team!
108
 
109
  I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
110
 
111
- If you're able and willing to contribute, it'd be most gratefully received and will help me to keep providing models, and work on new AI projects.
112
 
113
- Donaters will get priority support on any and all AI/LLM/model questions, plus other benefits.
114
 
115
  * Patreon: https://patreon.com/TheBlokeAI
116
  * Ko-Fi: https://ko-fi.com/TheBlokeAI
117
 
118
- **Patreon special mentions**: Aemon Algiz; Talal Aujan; Jonathan Leane; Illia Dulskyi; Khalefa Al-Ahmad;
119
- senxiiz. Thank you all, and to all my other generous patrons and donaters.
 
120
  <!-- footer end -->
121
 
122
  # Original model card: OpenAccess AI Collective's Hippogriff 30B Chat
@@ -126,7 +129,7 @@ senxiiz. Thank you all, and to all my other generous patrons and donaters.
126
  [<img src="https://huggingface.co/openaccess-ai-collective/hippogriff-30b-chat/resolve/main/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
127
 
128
 
129
- Hippogriff 30B Chat is an experiment that builds on Manticore with new datasets, while removing a few more instruction and chat datasets. It also includes a de-duped subset of the Pygmalion dataset. It also removes all Alpaca style prompts using `###` in favor of
130
  chat only style prompts using `USER:`,`ASSISTANT:` as well as [pygmalion/metharme prompting](https://huggingface.co/PygmalionAI/metharme-7b#prompting) using `<|system|>, <|user|> and <|model|>` tokens.
131
 
132
  Questions, comments, feedback, looking to donate, or want to help? Reach out on our [Discord](https://discord.gg/EqrvvehG) or email [wing@openaccessaicollective.org](mailto:wing@openaccessaicollective.org)
@@ -137,13 +140,13 @@ Hippogriff 30B Chat is a Llama 30B model fine-tuned on the following datasets
137
 
138
  - OpenAssistant/oasst1 - cleaned dataset, similar to Guanaco
139
  - synthetic jokes generation and explanation derived from reddit jokes dataset
140
- - synthetic prose generation and rewriting self-chat
141
  - Q&A based on provided context
142
  - self instruct augmented logic_inference_oa
143
  - de-duped pygmalion dataset, filtered down to RP data, cleaned, english only, 25%
144
- - [riddle_sense](https://huggingface.co/datasets/riddle_sense) - instruct augmented
145
  - hellaswag, updated for detailed explanations w 30K+ rows
146
- - [gsm8k](https://huggingface.co/datasets/gsm8k) - instruct augmented
147
  - [ewof/code-alpaca-instruct-unfiltered](https://huggingface.co/datasets/ewof/code-alpaca-instruct-unfiltered) synthetic self chat dataset derived from about 1000 rows
148
  - [subset of QingyiSi/Alpaca-CoT for roleplay and CoT](https://huggingface.co/QingyiSi/Alpaca-CoT)
149
  - [GPTeacher-General-Instruct](https://huggingface.co/datasets/teknium/GPTeacher-General-Instruct)
@@ -169,22 +172,22 @@ A Spaces demo is not provided for this release due to 30B models currently not f
169
 
170
  ## Build
171
 
172
- Hippogriff was built with [Axolotl](https://github.com/OpenAccess-AI-Collective/axolotl) on 8xA100 80GB
173
- - 1.5 epochs taking approximately 12 hours. No further epochs will be released.
174
 
175
  ## Bias, Risks, and Limitations
176
 
177
  Hippogriff has not been aligned to human preferences with techniques like RLHF or deployed with in-the-loop filtering of responses like ChatGPT, so the model can produce problematic outputs (especially when prompted to do so).
178
  Hippogriff was fine-tuned from the base model LlaMa 30B, please refer to its model card's Limitations Section for relevant information.
179
 
180
- The core functionality of Llama is taking a string of text and predicting the next token. While language models are widely used for tasks other than this,
181
- there are a lot of unknowns with this work. When prompting Hippogriff it is important to remember that the statistically most likely next token is often
182
  not the token that produces the most "accurate" text. Never depend upon Hippogriff to produce factually accurate output.
183
 
184
  Llama was trained on the Pile, a dataset known to contain profanity, lewd, and otherwise abrasive language. Depending upon use case Hippogriff may produce
185
  socially unacceptable text. See Sections 5 and 6 of the Pile paper for a more detailed analysis of the biases in the Pile.
186
 
187
- As with all language models, it is hard to predict in advance how Hippogriff will respond to particular prompts and offensive content may occur without warning.
188
  We recommend having a human curate or filter the outputs before releasing them, both to censor undesirable content and to improve the quality of the results.
189
 
190
  ## Examples
@@ -193,7 +196,7 @@ TBD
193
 
194
  ```
195
  USER: write an original short story of at least 300 words in the style of Zora Neale Hurston. Incorporate the following literary devices: Euphemism and Foreshadowing.
196
- ASSISTANT: Once upon a time, there was a black girl named Rose, whose skin shined like a pearl beneath the hot sun. She lived with her mama and daddy on their farm, where they grew crops to feed the family and neighbors. On this particular day, Rose felt a tinge of excitement as she watched her parents work diligently in their fields. 'It's almost harvest season!' she thought, imagining the delicious meals that awaited them. But, soon after, it became apparent that something sinister lurked amidst the rows of fresh produce.
197
 
198
  As her father hoed around a vine-covered cucumber plant, he unearthed a peculiar rock. It was the size of an acorn but had a strange shape, unlike anything he or his wife had ever seen. "What have we here?" He said, turning to his wife with curiosity and excitement. She examined the rock for herself and declared, "Why, it looks like a piece of coal!" But as both mother and daughter looked closer, it seemed to be far more than a mere piece of coal. A sense of unease crept up Rose's spine when she noticed that the rock resembled the dark face of a baby in the womb. As Rose tried to suppress a scream, her mother gasped, "Sweet Jesus! Let's bury this thing before anyone sees!" And so, they quickly hid the stone back underground, hoping that nobody would ever discover their secret.
199
  ```
 
23
  </div>
24
  <div style="display: flex; justify-content: space-between; width: 100%;">
25
  <div style="display: flex; flex-direction: column; align-items: flex-start;">
26
+ <p><a href="https://discord.gg/Jq4vkcDakD">Chat & support: my new Discord server</a></p>
27
  </div>
28
  <div style="display: flex; flex-direction: column; align-items: flex-end;">
29
  <p><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
 
100
  <!-- footer start -->
101
  ## Discord
102
 
103
+ For further support, and discussions on these models and AI in general, join us at:
104
+
105
+ [TheBloke AI's Discord server](https://discord.gg/Jq4vkcDakD)
106
 
107
  ## Thanks, and how to contribute.
108
 
 
110
 
111
  I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
112
 
113
+ If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
114
 
115
+ Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
116
 
117
  * Patreon: https://patreon.com/TheBlokeAI
118
  * Ko-Fi: https://ko-fi.com/TheBlokeAI
119
 
120
+ **Patreon special mentions**: Aemon Algiz, Dmitriy Samsonov, Nathan LeClaire, Trenton Dambrowitz, Mano Prime, David Flickinger, vamX, Nikolai Manek, senxiiz, Khalefa Al-Ahmad, Illia Dulskyi, Jonathan Leane, Talal Aujan, V. Lukas, Joseph William Delisle, Pyrater, Oscar Rangel, Lone Striker, Luke Pendergrass, Eugene Pentland, Sebastain Graf, Johann-Peter Hartman.
121
+
122
+ Thank you to all my generous patrons and donaters!
123
  <!-- footer end -->
124
 
125
  # Original model card: OpenAccess AI Collective's Hippogriff 30B Chat
 
129
  [<img src="https://huggingface.co/openaccess-ai-collective/hippogriff-30b-chat/resolve/main/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
130
 
131
 
132
+ Hippogriff 30B Chat is an experiment that builds on Manticore with new datasets, while removing a few more instruction and chat datasets. It also includes a de-duped subset of the Pygmalion dataset. It also removes all Alpaca style prompts using `###` in favor of
133
  chat only style prompts using `USER:`,`ASSISTANT:` as well as [pygmalion/metharme prompting](https://huggingface.co/PygmalionAI/metharme-7b#prompting) using `<|system|>, <|user|> and <|model|>` tokens.
134
 
135
  Questions, comments, feedback, looking to donate, or want to help? Reach out on our [Discord](https://discord.gg/EqrvvehG) or email [wing@openaccessaicollective.org](mailto:wing@openaccessaicollective.org)
 
140
 
141
  - OpenAssistant/oasst1 - cleaned dataset, similar to Guanaco
142
  - synthetic jokes generation and explanation derived from reddit jokes dataset
143
+ - synthetic prose generation and rewriting self-chat
144
  - Q&A based on provided context
145
  - self instruct augmented logic_inference_oa
146
  - de-duped pygmalion dataset, filtered down to RP data, cleaned, english only, 25%
147
+ - [riddle_sense](https://huggingface.co/datasets/riddle_sense) - instruct augmented
148
  - hellaswag, updated for detailed explanations w 30K+ rows
149
+ - [gsm8k](https://huggingface.co/datasets/gsm8k) - instruct augmented
150
  - [ewof/code-alpaca-instruct-unfiltered](https://huggingface.co/datasets/ewof/code-alpaca-instruct-unfiltered) synthetic self chat dataset derived from about 1000 rows
151
  - [subset of QingyiSi/Alpaca-CoT for roleplay and CoT](https://huggingface.co/QingyiSi/Alpaca-CoT)
152
  - [GPTeacher-General-Instruct](https://huggingface.co/datasets/teknium/GPTeacher-General-Instruct)
 
172
 
173
  ## Build
174
 
175
+ Hippogriff was built with [Axolotl](https://github.com/OpenAccess-AI-Collective/axolotl) on 8xA100 80GB
176
+ - 1.5 epochs taking approximately 12 hours. No further epochs will be released.
177
 
178
  ## Bias, Risks, and Limitations
179
 
180
  Hippogriff has not been aligned to human preferences with techniques like RLHF or deployed with in-the-loop filtering of responses like ChatGPT, so the model can produce problematic outputs (especially when prompted to do so).
181
  Hippogriff was fine-tuned from the base model LlaMa 30B, please refer to its model card's Limitations Section for relevant information.
182
 
183
+ The core functionality of Llama is taking a string of text and predicting the next token. While language models are widely used for tasks other than this,
184
+ there are a lot of unknowns with this work. When prompting Hippogriff it is important to remember that the statistically most likely next token is often
185
  not the token that produces the most "accurate" text. Never depend upon Hippogriff to produce factually accurate output.
186
 
187
  Llama was trained on the Pile, a dataset known to contain profanity, lewd, and otherwise abrasive language. Depending upon use case Hippogriff may produce
188
  socially unacceptable text. See Sections 5 and 6 of the Pile paper for a more detailed analysis of the biases in the Pile.
189
 
190
+ As with all language models, it is hard to predict in advance how Hippogriff will respond to particular prompts and offensive content may occur without warning.
191
  We recommend having a human curate or filter the outputs before releasing them, both to censor undesirable content and to improve the quality of the results.
192
 
193
  ## Examples
 
196
 
197
  ```
198
  USER: write an original short story of at least 300 words in the style of Zora Neale Hurston. Incorporate the following literary devices: Euphemism and Foreshadowing.
199
+ ASSISTANT: Once upon a time, there was a black girl named Rose, whose skin shined like a pearl beneath the hot sun. She lived with her mama and daddy on their farm, where they grew crops to feed the family and neighbors. On this particular day, Rose felt a tinge of excitement as she watched her parents work diligently in their fields. 'It's almost harvest season!' she thought, imagining the delicious meals that awaited them. But, soon after, it became apparent that something sinister lurked amidst the rows of fresh produce.
200
 
201
  As her father hoed around a vine-covered cucumber plant, he unearthed a peculiar rock. It was the size of an acorn but had a strange shape, unlike anything he or his wife had ever seen. "What have we here?" He said, turning to his wife with curiosity and excitement. She examined the rock for herself and declared, "Why, it looks like a piece of coal!" But as both mother and daughter looked closer, it seemed to be far more than a mere piece of coal. A sense of unease crept up Rose's spine when she noticed that the rock resembled the dark face of a baby in the womb. As Rose tried to suppress a scream, her mother gasped, "Sweet Jesus! Let's bury this thing before anyone sees!" And so, they quickly hid the stone back underground, hoping that nobody would ever discover their secret.
202
  ```