johannhartmann
commited on
Upload folder using huggingface_hub
Browse files- .gitattributes +15 -0
- README.md +324 -0
- llama-3-sauerkrautlm-8b-instruct.Q2_K.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.Q3_K_L.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.Q3_K_M.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.Q3_K_S.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.Q4_0.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.Q4_1.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.Q4_K_M.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.Q4_K_S.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.Q5_0.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.Q5_1.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.Q5_K_M.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.Q5_K_S.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.Q6_K.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.Q8_0.gguf +3 -0
- llama-3-sauerkrautlm-8b-instruct.gguf +3 -0
.gitattributes
CHANGED
@@ -33,3 +33,18 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
llama-3-sauerkrautlm-8b-instruct.Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
|
37 |
+
llama-3-sauerkrautlm-8b-instruct.Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
|
38 |
+
llama-3-sauerkrautlm-8b-instruct.Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
39 |
+
llama-3-sauerkrautlm-8b-instruct.Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
40 |
+
llama-3-sauerkrautlm-8b-instruct.Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
|
41 |
+
llama-3-sauerkrautlm-8b-instruct.Q4_1.gguf filter=lfs diff=lfs merge=lfs -text
|
42 |
+
llama-3-sauerkrautlm-8b-instruct.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
43 |
+
llama-3-sauerkrautlm-8b-instruct.Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
44 |
+
llama-3-sauerkrautlm-8b-instruct.Q5_0.gguf filter=lfs diff=lfs merge=lfs -text
|
45 |
+
llama-3-sauerkrautlm-8b-instruct.Q5_1.gguf filter=lfs diff=lfs merge=lfs -text
|
46 |
+
llama-3-sauerkrautlm-8b-instruct.Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
47 |
+
llama-3-sauerkrautlm-8b-instruct.Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
48 |
+
llama-3-sauerkrautlm-8b-instruct.Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
|
49 |
+
llama-3-sauerkrautlm-8b-instruct.Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
|
50 |
+
llama-3-sauerkrautlm-8b-instruct.gguf filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,324 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- de
|
4 |
+
- en
|
5 |
+
license: other
|
6 |
+
tags:
|
7 |
+
- two stage dpo
|
8 |
+
- dpo
|
9 |
+
- gguf
|
10 |
+
license_name: llama3
|
11 |
+
license_link: LICENSE
|
12 |
+
extra_gated_prompt: "### META LLAMA 3 COMMUNITY LICENSE AGREEMENT\nMeta Llama 3 Version\
|
13 |
+
\ Release Date: April 18, 2024\n\"Agreement\" means the terms and conditions for\
|
14 |
+
\ use, reproduction, distribution and modification of the Llama Materials set forth\
|
15 |
+
\ herein.\n\"Documentation\" means the specifications, manuals and documentation\
|
16 |
+
\ accompanying Meta Llama 3 distributed by Meta at https://llama.meta.com/get-started/.\n\
|
17 |
+
\"Licensee\" or \"you\" means you, or your employer or any other person or entity\
|
18 |
+
\ (if you are entering into this Agreement on such person or entity’s behalf), of\
|
19 |
+
\ the age required under applicable laws, rules or regulations to provide legal\
|
20 |
+
\ consent and that has legal authority to bind your employer or such other person\
|
21 |
+
\ or entity if you are entering in this Agreement on their behalf.\n\"Meta Llama\
|
22 |
+
\ 3\" means the foundational large language models and software and algorithms,\
|
23 |
+
\ including machine-learning model code, trained model weights, inference-enabling\
|
24 |
+
\ code, training-enabling code, fine-tuning enabling code and other elements of\
|
25 |
+
\ the foregoing distributed by Meta at https://llama.meta.com/llama-downloads.\n\
|
26 |
+
\"Llama Materials\" means, collectively, Meta’s proprietary Meta Llama 3 and Documentation\
|
27 |
+
\ (and any portion thereof) made available under this Agreement.\n\"Meta\" or \"\
|
28 |
+
we\" means Meta Platforms Ireland Limited (if you are located in or, if you are\
|
29 |
+
\ an entity, your principal place of business is in the EEA or Switzerland) and\
|
30 |
+
\ Meta Platforms, Inc. (if you are located outside of the EEA or Switzerland).\n\
|
31 |
+
\ \n1. License Rights and Redistribution.\na. Grant of Rights. You are granted\
|
32 |
+
\ a non-exclusive, worldwide, non-transferable and royalty-free limited license\
|
33 |
+
\ under Meta’s intellectual property or other rights owned by Meta embodied in the\
|
34 |
+
\ Llama Materials to use, reproduce, distribute, copy, create derivative works of,\
|
35 |
+
\ and make modifications to the Llama Materials.\nb. Redistribution and Use.\ni.\
|
36 |
+
\ If you distribute or make available the Llama Materials (or any derivative works\
|
37 |
+
\ thereof), or a product or service that uses any of them, including another AI\
|
38 |
+
\ model, you shall (A) provide a copy of this Agreement with any such Llama Materials;\
|
39 |
+
\ and (B) prominently display “Built with Meta Llama 3” on a related website, user\
|
40 |
+
\ interface, blogpost, about page, or product documentation. If you use the Llama\
|
41 |
+
\ Materials to create, train, fine tune, or otherwise improve an AI model, which\
|
42 |
+
\ is distributed or made available, you shall also include “Llama 3” at the beginning\
|
43 |
+
\ of any such AI model name.\nii. If you receive Llama Materials, or any derivative\
|
44 |
+
\ works thereof, from a Licensee as part of an integrated end user product, then\
|
45 |
+
\ Section 2 of this Agreement will not apply to you.\niii. You must retain in all\
|
46 |
+
\ copies of the Llama Materials that you distribute the following attribution notice\
|
47 |
+
\ within a “Notice” text file distributed as a part of such copies: “Meta Llama\
|
48 |
+
\ 3 is licensed under the Meta Llama 3 Community License, Copyright © Meta Platforms,\
|
49 |
+
\ Inc. All Rights Reserved.”\niv. Your use of the Llama Materials must comply with\
|
50 |
+
\ applicable laws and regulations (including trade compliance laws and regulations)\
|
51 |
+
\ and adhere to the Acceptable Use Policy for the Llama Materials (available at\
|
52 |
+
\ https://llama.meta.com/llama3/use-policy), which is hereby incorporated by reference\
|
53 |
+
\ into this Agreement.\nv. You will not use the Llama Materials or any output or\
|
54 |
+
\ results of the Llama Materials to improve any other large language model (excluding\
|
55 |
+
\ Meta Llama 3 or derivative works thereof).\n2. Additional Commercial Terms. If,\
|
56 |
+
\ on the Meta Llama 3 version release date, the monthly active users of the products\
|
57 |
+
\ or services made available by or for Licensee, or Licensee’s affiliates, is greater\
|
58 |
+
\ than 700 million monthly active users in the preceding calendar month, you must\
|
59 |
+
\ request a license from Meta, which Meta may grant to you in its sole discretion,\
|
60 |
+
\ and you are not authorized to exercise any of the rights under this Agreement\
|
61 |
+
\ unless or until Meta otherwise expressly grants you such rights.\n3. Disclaimer\
|
62 |
+
\ of Warranty. UNLESS REQUIRED BY APPLICABLE LAW, THE LLAMA MATERIALS AND ANY OUTPUT\
|
63 |
+
\ AND RESULTS THEREFROM ARE PROVIDED ON AN “AS IS” BASIS, WITHOUT WARRANTIES OF\
|
64 |
+
\ ANY KIND, AND META DISCLAIMS ALL WARRANTIES OF ANY KIND, BOTH EXPRESS AND IMPLIED,\
|
65 |
+
\ INCLUDING, WITHOUT LIMITATION, ANY WARRANTIES OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY,\
|
66 |
+
\ OR FITNESS FOR A PARTICULAR PURPOSE. YOU ARE SOLELY RESPONSIBLE FOR DETERMINING\
|
67 |
+
\ THE APPROPRIATENESS OF USING OR REDISTRIBUTING THE LLAMA MATERIALS AND ASSUME\
|
68 |
+
\ ANY RISKS ASSOCIATED WITH YOUR USE OF THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS.\n\
|
69 |
+
4. Limitation of Liability. IN NO EVENT WILL META OR ITS AFFILIATES BE LIABLE UNDER\
|
70 |
+
\ ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, TORT, NEGLIGENCE, PRODUCTS LIABILITY,\
|
71 |
+
\ OR OTHERWISE, ARISING OUT OF THIS AGREEMENT, FOR ANY LOST PROFITS OR ANY INDIRECT,\
|
72 |
+
\ SPECIAL, CONSEQUENTIAL, INCIDENTAL, EXEMPLARY OR PUNITIVE DAMAGES, EVEN IF META\
|
73 |
+
\ OR ITS AFFILIATES HAVE BEEN ADVISED OF THE POSSIBILITY OF ANY OF THE FOREGOING.\n\
|
74 |
+
5. Intellectual Property.\na. No trademark licenses are granted under this Agreement,\
|
75 |
+
\ and in connection with the Llama Materials, neither Meta nor Licensee may use\
|
76 |
+
\ any name or mark owned by or associated with the other or any of its affiliates,\
|
77 |
+
\ except as required for reasonable and customary use in describing and redistributing\
|
78 |
+
\ the Llama Materials or as set forth in this Section 5(a). Meta hereby grants you\
|
79 |
+
\ a license to use “Llama 3” (the “Mark”) solely as required to comply with the\
|
80 |
+
\ last sentence of Section 1.b.i. You will comply with Meta’s brand guidelines (currently\
|
81 |
+
\ accessible at https://about.meta.com/brand/resources/meta/company-brand/ ). All\
|
82 |
+
\ goodwill arising out of your use of the Mark will inure to the benefit of Meta.\n\
|
83 |
+
b. Subject to Meta’s ownership of Llama Materials and derivatives made by or for\
|
84 |
+
\ Meta, with respect to any derivative works and modifications of the Llama Materials\
|
85 |
+
\ that are made by you, as between you and Meta, you are and will be the owner of\
|
86 |
+
\ such derivative works and modifications.\nc. If you institute litigation or other\
|
87 |
+
\ proceedings against Meta or any entity (including a cross-claim or counterclaim\
|
88 |
+
\ in a lawsuit) alleging that the Llama Materials or Meta Llama 3 outputs or results,\
|
89 |
+
\ or any portion of any of the foregoing, constitutes infringement of intellectual\
|
90 |
+
\ property or other rights owned or licensable by you, then any licenses granted\
|
91 |
+
\ to you under this Agreement shall terminate as of the date such litigation or\
|
92 |
+
\ claim is filed or instituted. You will indemnify and hold harmless Meta from and\
|
93 |
+
\ against any claim by any third party arising out of or related to your use or\
|
94 |
+
\ distribution of the Llama Materials.\n6. Term and Termination. The term of this\
|
95 |
+
\ Agreement will commence upon your acceptance of this Agreement or access to the\
|
96 |
+
\ Llama Materials and will continue in full force and effect until terminated in\
|
97 |
+
\ accordance with the terms and conditions herein. Meta may terminate this Agreement\
|
98 |
+
\ if you are in breach of any term or condition of this Agreement. Upon termination\
|
99 |
+
\ of this Agreement, you shall delete and cease use of the Llama Materials. Sections\
|
100 |
+
\ 3, 4 and 7 shall survive the termination of this Agreement.\n7. Governing Law\
|
101 |
+
\ and Jurisdiction. This Agreement will be governed and construed under the laws\
|
102 |
+
\ of the State of California without regard to choice of law principles, and the\
|
103 |
+
\ UN Convention on Contracts for the International Sale of Goods does not apply\
|
104 |
+
\ to this Agreement. The courts of California shall have exclusive jurisdiction\
|
105 |
+
\ of any dispute arising out of this Agreement.\n### Meta Llama 3 Acceptable Use\
|
106 |
+
\ Policy\nMeta is committed to promoting safe and fair use of its tools and features,\
|
107 |
+
\ including Meta Llama 3. If you access or use Meta Llama 3, you agree to this Acceptable\
|
108 |
+
\ Use Policy (“Policy”). The most recent copy of this policy can be found at [https://llama.meta.com/llama3/use-policy](https://llama.meta.com/llama3/use-policy)\n\
|
109 |
+
#### Prohibited Uses\nWe want everyone to use Meta Llama 3 safely and responsibly.\
|
110 |
+
\ You agree you will not use, or allow others to use, Meta Llama 3 to: 1. Violate\
|
111 |
+
\ the law or others’ rights, including to:\n 1. Engage in, promote, generate,\
|
112 |
+
\ contribute to, encourage, plan, incite, or further illegal or unlawful activity\
|
113 |
+
\ or content, such as:\n 1. Violence or terrorism\n 2. Exploitation\
|
114 |
+
\ or harm to children, including the solicitation, creation, acquisition, or dissemination\
|
115 |
+
\ of child exploitative content or failure to report Child Sexual Abuse Material\n\
|
116 |
+
\ 3. Human trafficking, exploitation, and sexual violence\n 4. The\
|
117 |
+
\ illegal distribution of information or materials to minors, including obscene\
|
118 |
+
\ materials, or failure to employ legally required age-gating in connection with\
|
119 |
+
\ such information or materials.\n 5. Sexual solicitation\n 6. Any\
|
120 |
+
\ other criminal activity\n 2. Engage in, promote, incite, or facilitate the\
|
121 |
+
\ harassment, abuse, threatening, or bullying of individuals or groups of individuals\n\
|
122 |
+
\ 3. Engage in, promote, incite, or facilitate discrimination or other unlawful\
|
123 |
+
\ or harmful conduct in the provision of employment, employment benefits, credit,\
|
124 |
+
\ housing, other economic benefits, or other essential goods and services\n 4.\
|
125 |
+
\ Engage in the unauthorized or unlicensed practice of any profession including,\
|
126 |
+
\ but not limited to, financial, legal, medical/health, or related professional\
|
127 |
+
\ practices\n 5. Collect, process, disclose, generate, or infer health, demographic,\
|
128 |
+
\ or other sensitive personal or private information about individuals without rights\
|
129 |
+
\ and consents required by applicable laws\n 6. Engage in or facilitate any action\
|
130 |
+
\ or generate any content that infringes, misappropriates, or otherwise violates\
|
131 |
+
\ any third-party rights, including the outputs or results of any products or services\
|
132 |
+
\ using the Llama Materials\n 7. Create, generate, or facilitate the creation\
|
133 |
+
\ of malicious code, malware, computer viruses or do anything else that could disable,\
|
134 |
+
\ overburden, interfere with or impair the proper working, integrity, operation\
|
135 |
+
\ or appearance of a website or computer system\n2. Engage in, promote, incite,\
|
136 |
+
\ facilitate, or assist in the planning or development of activities that present\
|
137 |
+
\ a risk of death or bodily harm to individuals, including use of Meta Llama 3 related\
|
138 |
+
\ to the following:\n 1. Military, warfare, nuclear industries or applications,\
|
139 |
+
\ espionage, use for materials or activities that are subject to the International\
|
140 |
+
\ Traffic Arms Regulations (ITAR) maintained by the United States Department of\
|
141 |
+
\ State\n 2. Guns and illegal weapons (including weapon development)\n 3.\
|
142 |
+
\ Illegal drugs and regulated/controlled substances\n 4. Operation of critical\
|
143 |
+
\ infrastructure, transportation technologies, or heavy machinery\n 5. Self-harm\
|
144 |
+
\ or harm to others, including suicide, cutting, and eating disorders\n 6. Any\
|
145 |
+
\ content intended to incite or promote violence, abuse, or any infliction of bodily\
|
146 |
+
\ harm to an individual\n3. Intentionally deceive or mislead others, including use\
|
147 |
+
\ of Meta Llama 3 related to the following:\n 1. Generating, promoting, or furthering\
|
148 |
+
\ fraud or the creation or promotion of disinformation\n 2. Generating, promoting,\
|
149 |
+
\ or furthering defamatory content, including the creation of defamatory statements,\
|
150 |
+
\ images, or other content\n 3. Generating, promoting, or further distributing\
|
151 |
+
\ spam\n 4. Impersonating another individual without consent, authorization,\
|
152 |
+
\ or legal right\n 5. Representing that the use of Meta Llama 3 or outputs are\
|
153 |
+
\ human-generated\n 6. Generating or facilitating false online engagement, including\
|
154 |
+
\ fake reviews and other means of fake online engagement\n4. Fail to appropriately\
|
155 |
+
\ disclose to end users any known dangers of your AI system\nPlease report any violation\
|
156 |
+
\ of this Policy, software “bug,” or other problems that could lead to a violation\
|
157 |
+
\ of this Policy through one of the following means:\n * Reporting issues with\
|
158 |
+
\ the model: [https://github.com/meta-llama/llama3](https://github.com/meta-llama/llama3)\n\
|
159 |
+
\ * Reporting risky content generated by the model:\n developers.facebook.com/llama_output_feedback\n\
|
160 |
+
\ * Reporting bugs and security concerns: facebook.com/whitehat/info\n * Reporting\
|
161 |
+
\ violations of the Acceptable Use Policy or unlicensed uses of Meta Llama 3: LlamaUseReport@meta.com"
|
162 |
+
extra_gated_fields:
|
163 |
+
First Name: text
|
164 |
+
Last Name: text
|
165 |
+
Date of birth: date_picker
|
166 |
+
Country: country
|
167 |
+
Affiliation: text
|
168 |
+
geo: ip_location
|
169 |
+
? By clicking Submit below I accept the terms of the license and acknowledge that
|
170 |
+
the information I provide will be collected stored processed and shared in accordance
|
171 |
+
with the Meta Privacy Policy
|
172 |
+
: checkbox
|
173 |
+
extra_gated_description: The information you provide will be collected, stored, processed
|
174 |
+
and shared in accordance with the [Meta Privacy Policy](https://www.facebook.com/privacy/policy/).
|
175 |
+
extra_gated_button_content: Submit
|
176 |
+
---
|
177 |
+
|
178 |
+
|
179 |
+
![SauerkrautLM](https://vago-solutions.ai/wp-content/uploads/2024/04/Llama3-Pic.png "Llama-3-SauerkrautLM-8b-Instruct")
|
180 |
+
## VAGO solutions Llama-3-SauerkrautLM-8b-Instruct
|
181 |
+
Introducing **Llama-3-SauerkrautLM-8b-Instruct** – our Sauerkraut version of the powerful [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct)!
|
182 |
+
|
183 |
+
The model **Llama-3-SauerkrautLM-8b-Instruct** is a **joint effort** between **VAGO Solutions** and **Hyperspace.ai.**
|
184 |
+
|
185 |
+
- Aligned with **DPO**
|
186 |
+
|
187 |
+
# Table of Contents
|
188 |
+
1. [Overview of all Llama-3-SauerkrautLM-8b-Instruct](#all-Llama-3-SauerkrautLM-8b-Instruct)
|
189 |
+
2. [Model Details](#model-details)
|
190 |
+
- [Prompt template](#prompt-template)
|
191 |
+
- [Training procedure](#proceed-of-the-training)
|
192 |
+
3. [Evaluation](#evaluation)
|
193 |
+
5. [Disclaimer](#disclaimer)
|
194 |
+
6. [Contact](#contact)
|
195 |
+
7. [Collaborations](#collaborations)
|
196 |
+
8. [Acknowledgement](#acknowledgement)
|
197 |
+
|
198 |
+
|
199 |
+
## All SauerkrautLM-llama-3-8B-Instruct
|
200 |
+
|
201 |
+
| Model | HF | EXL2 | GGUF | AWQ |
|
202 |
+
|-------|-------|-------|-------|-------|
|
203 |
+
| Llama-3-SauerkrautLM-8b-Instruct | [Link](https://huggingface.co/VAGOsolutions/Llama-3-SauerkrautLM-8b-Instruct) | [Link](https://huggingface.co/bartowski/Llama-3-SauerkrautLM-8b-Instruct-exl2) | [Link](https://huggingface.co/bartowski/Llama-3-SauerkrautLM-8b-Instruct-GGUF) | coming soon |
|
204 |
+
|
205 |
+
## Model Details
|
206 |
+
**SauerkrautLM-llama-3-8B-Instruct**
|
207 |
+
- **Model Type:** Llama-3-SauerkrautLM-8b-Instruct is a finetuned Model based on [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct)
|
208 |
+
- **Language(s):** German, English
|
209 |
+
- **License:** [meta-llama](https://llama.meta.com/llama3/license)
|
210 |
+
- **Contact:** [VAGO solutions](https://vago-solutions.ai), [Hyperspace.ai](https://hyperspace.computer/)
|
211 |
+
|
212 |
+
### Training procedure:
|
213 |
+
- We trained this model in a two staged DPO Fine-Tuning for 1 epoch with 70k data and another epoch with 20k data.
|
214 |
+
- LaserRMT version coming soon
|
215 |
+
|
216 |
+
**We improved the model's capabilities noticably by feeding it with curated German data.**
|
217 |
+
|
218 |
+
|
219 |
+
### Prompt Template:
|
220 |
+
|
221 |
+
**English:**
|
222 |
+
```
|
223 |
+
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
|
224 |
+
|
225 |
+
You are a helpful AI assistant.<|eot_id|><|start_header_id|>user<|end_header_id|>
|
226 |
+
|
227 |
+
Input<|eot_id|><|start_header_id|>assistant<|end_header_id|>
|
228 |
+
|
229 |
+
```
|
230 |
+
|
231 |
+
**German:**
|
232 |
+
```
|
233 |
+
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
|
234 |
+
|
235 |
+
Du bist ein freundlicher und hilfreicher deutscher KI-Assistent.<|eot_id|><|start_header_id|>user<|end_header_id|>
|
236 |
+
|
237 |
+
Input<|eot_id|><|start_header_id|>assistant<|end_header_id|>
|
238 |
+
|
239 |
+
```
|
240 |
+
### Example output of german language:
|
241 |
+
|
242 |
+
```
|
243 |
+
|
244 |
+
Prompt: Erkläre mir die Vorteile für das Fine-Tuning von Meta's LLama Modellen.
|
245 |
+
|
246 |
+
Antwort:
|
247 |
+
Sehr gern!
|
248 |
+
|
249 |
+
Fine-tuning von Meta's LLaMA-Modellen ist eine sehr effektive Möglichkeit, um Ihre Anwendung oder Ihr Projekt mit einer hochentwickelten Sprachverarbeitung auszustatten.
|
250 |
+
|
251 |
+
Die Vorteile des Fine-Tunings von LLaMA-Modellen sind:
|
252 |
+
|
253 |
+
Adaptation an spezielle Aufgaben: Durch das Fine-Tuning können Sie die Modelle auf Ihre speziellen Bedürfnisse und Anforderungen anpassen.
|
254 |
+
Dies ermöglicht es Ihnen, die Leistung der Modelle in Ihren Anwendungen zu verbessern.
|
255 |
+
|
256 |
+
Verbesserte Genauigkeit: Die fine-tuned Modelle lernen, sich an die spezifischen Merkmale Ihres Datenbestands anzupassen, was zu einer besseren Genauigkeit bei der Textklassifizierung, -generierung, -summarisierung usw. führt.
|
257 |
+
|
258 |
+
Anpassung an Branche oder Domain: Wenn Sie beispielsweise ein Medizinprojekt haben, können Sie die Modelle auf medizinische Fachterminologie und -konzepte trainieren, um eine höhere Präzision bei der Analyse medizinischer Texte zu erzielen.
|
259 |
+
|
260 |
+
Kosten-Effektivität: Im Vergleich zum Training neuer Modelle von Grund auf kann das Fine-Tuning viel kosteneffektiver sein, da Sie nur einen kleinen Teil des Trainingsdatums benötigen, um die Modelle an Ihre spezifischen Anforderungen anzupassen.
|
261 |
+
|
262 |
+
Schnellere Entwicklung: Das Fine-Tuning ermöglicht es Entwicklern, schnell neue Funktionen und Features in ihre Anwendungen einzufügen, ohne dass sie lange warten müssen, bis neue Modelle trainiert werden.
|
263 |
+
|
264 |
+
Bessere Interpretierbarkeit: Da die fine-tuned Modelle auf Ihre spezifischen Daten bestimmt wurden, können Sie leichter verstehen, warum bestimmte Entscheidungen getroffen werden, was wichtig ist, wenn es um Transparenz und Verantwortlichkeit geht.
|
265 |
+
|
266 |
+
Insgesamt bietet das Fine-Tuning von LLaMA-Modellen eine flexible und effektive Möglichkeit, um Ihre Anwendungen und Projekte durch die Integration von fortschrittlichen Sprachmodellen zu verbessern.
|
267 |
+
```
|
268 |
+
|
269 |
+
## Evaluation
|
270 |
+
|
271 |
+
**Open LLM Leaderboard:**
|
272 |
+
|
273 |
+
evaluated with lm-evaluation-benchmark-harness 0.4.2
|
274 |
+
| Metric | Value |
|
275 |
+
|-----------------------|---------------------------|
|
276 |
+
| Avg. | **74.57** |
|
277 |
+
| ARC (25-shot) | 74.66 |
|
278 |
+
| HellaSwag (10-shot) | 89.60 |
|
279 |
+
| MMLU (5-shot) | 66.55 |
|
280 |
+
| TruthfulQA (0-shot) | 66.32 |
|
281 |
+
| Winogrande (5-shot) | 80.98 |
|
282 |
+
| GSM8K (5-shot) | 69.29 |
|
283 |
+
|
284 |
+
|
285 |
+
**MT-Bench English**
|
286 |
+
|
287 |
+
```
|
288 |
+
########## First turn ##########
|
289 |
+
score
|
290 |
+
model turn
|
291 |
+
Llama-3-SauerkrautLM-8b-Instruct 1 8.15625
|
292 |
+
|
293 |
+
########## Second turn ##########
|
294 |
+
score
|
295 |
+
model turn
|
296 |
+
Llama-3-SauerkrautLM-8b-Instruct 2 7.65
|
297 |
+
|
298 |
+
########## Average ##########
|
299 |
+
score
|
300 |
+
model
|
301 |
+
Llama-3-SauerkrautLM-8b-Instruct 7.903125 *
|
302 |
+
|
303 |
+
```
|
304 |
+
* due to specific instruction training the english MT-Bench score is slightly lower than the original LLama-3-8B-Instruct
|
305 |
+
|
306 |
+
|
307 |
+
**MT-Bench German**
|
308 |
+
coming soon
|
309 |
+
|
310 |
+
|
311 |
+
## Disclaimer
|
312 |
+
We must inform users that despite our best efforts in data cleansing, the possibility of uncensored content slipping through cannot be entirely ruled out.
|
313 |
+
However, we cannot guarantee consistently appropriate behavior. Therefore, if you encounter any issues or come across inappropriate content, we kindly request that you inform us through the contact information provided.
|
314 |
+
Additionally, it is essential to understand that the licensing of these models does not constitute legal advice. We are not held responsible for the actions of third parties who utilize our models.
|
315 |
+
|
316 |
+
## Contact
|
317 |
+
If you are interested in customized LLMs for business applications, please get in contact with us via our websites. We are also grateful for your feedback and suggestions.
|
318 |
+
|
319 |
+
## Collaborations
|
320 |
+
We are also keenly seeking support and investment for our startups, VAGO solutions and Hyperspace where we continuously advance the development of robust language models designed to address a diverse range of purposes and requirements. If the prospect of collaboratively navigating future challenges excites you, we warmly invite you to reach out to us at [VAGO solutions](https://vago-solutions.de/#Kontakt), [Hyperspace.computer](https://hyperspace.computer/)
|
321 |
+
|
322 |
+
## Acknowledgement
|
323 |
+
Many thanks to [Meta](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) for providing such valuable model to the Open-Source community.
|
324 |
+
Also many thanks to [bartowski](https://huggingface.co/bartowski) for super fast quantification of our Model in GGUF and EXL format.
|
llama-3-sauerkrautlm-8b-instruct.Q2_K.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:344bc4f17ce95690da47a86d2e5be171d30660e6498e9d7b846a5dcbb41f7353
|
3 |
+
size 3179131136
|
llama-3-sauerkrautlm-8b-instruct.Q3_K_L.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3b3d1c22a6f1f9f99212bb16a4f0b952df476c333c2a385cb35bda22d5259b87
|
3 |
+
size 4321956096
|
llama-3-sauerkrautlm-8b-instruct.Q3_K_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1ee255f8dbd084f7943b3279c0a62e33d7202d70fd87ca47d52a5c6fa20f3c3a
|
3 |
+
size 4018917632
|
llama-3-sauerkrautlm-8b-instruct.Q3_K_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f12394d475e19303b5134e8332511d44669009c408cd5f865d93f4c2fdf50bef
|
3 |
+
size 3664498944
|
llama-3-sauerkrautlm-8b-instruct.Q4_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d6ca2b00c40b782f833f9d017a012a9d6acaa96468ee451d3dfe94cd2522f7a3
|
3 |
+
size 4661211392
|
llama-3-sauerkrautlm-8b-instruct.Q4_1.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5ddb16ecf9e0c58bc50a094e3641d534956e0f1e0a68a3d88ef90c256a298fb0
|
3 |
+
size 5130252544
|
llama-3-sauerkrautlm-8b-instruct.Q4_K_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:08720097e2ce392fcc395612ad7f96f6ed3b7d61f8c9859acf6c7f15bbe4bfa8
|
3 |
+
size 4920733952
|
llama-3-sauerkrautlm-8b-instruct.Q4_K_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6c20f688a9781ffe5acdbef8b7c08e82d5cf7eb5c51b1bd59b117d01232d7187
|
3 |
+
size 4692668672
|
llama-3-sauerkrautlm-8b-instruct.Q5_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:01431c2517586525445d0519304e5ded9e840683ded9ebd46ba657bf1f3a8f80
|
3 |
+
size 5599293696
|
llama-3-sauerkrautlm-8b-instruct.Q5_1.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e2572118bcd2ef6ccc97a28666e02c4f33d0531f244516c05f93fd56f94be353
|
3 |
+
size 6068334848
|
llama-3-sauerkrautlm-8b-instruct.Q5_K_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:60b784153773a3b36b84a935154758b01d1f61ff36a180d69a38afe32fc8e11b
|
3 |
+
size 5732987136
|
llama-3-sauerkrautlm-8b-instruct.Q5_K_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ea11ed9ad17072ad40ddb05ba10a998b9d62d61a11089c8fa87fc8eef68121e8
|
3 |
+
size 5599293696
|
llama-3-sauerkrautlm-8b-instruct.Q6_K.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5f9a43c8bd31645fbd09f44b417c26d337ab872f03deed46daaf3f6f46604a41
|
3 |
+
size 6596006144
|
llama-3-sauerkrautlm-8b-instruct.Q8_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:62c7bde06540bfc6655c45c0bcf1e6dadb654031cd5399a7bede21d2b2837570
|
3 |
+
size 8540770560
|
llama-3-sauerkrautlm-8b-instruct.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6e11406045445e24d282db00c264fcfb7f139da18656bb0623c17e93e59731a0
|
3 |
+
size 16068890816
|