--- license: apache-2.0 language: - en - de library_name: transformers pipeline_tag: text-generation tags: - mistral - finetune - chatml - augmentation - german --- ![SauerkrautLM](https://vago-solutions.de/wp-content/uploads/2023/11/hero.png "SauerkrautLM-7b-HerO") ## VAGO solutions SauerkrautLM-7b-HerO Introducing **SauerkrautLM-7b-HerO** – the pinnacle of German language model technology! Crafted through the **merging** of **[Teknium's OpenHermes-2.5-Mistral-7B](https://huggingface.co/teknium/OpenHermes-2.5-Mistral-7B)** and **[Open-Orca's Mistral-7B-OpenOrca](https://huggingface.co/Open-Orca/Mistral-7B-OpenOrca)** and **uniquely fine-tuned with the Sauerkraut dataset.** SauerkrautLM-7b-HerO represents a breakthrough in language modeling, achieving an optimal balance between extensive German data and essential international sources. This ensures the model not only excels in understanding the nuances of the German language but also retains its global capabilities. Harnessing the innovative power of the **gradient SLERP method from MergeKit**, we've achieved a groundbreaking fusion of two of the most best performing 7B models based on the Mistral framework. This merge has allowed us to combine the best features of both models, creating an unparalleled synergy. Coupled with the German Sauerkraut dataset, which consists of a mix of augmented and translated data, we have successfully taught the English-speaking merged model the intricacies of the German language. This was achieved *without the typical loss of core competencies often associated with fine-tuning in another language of models previously trained mainly in English.* Our approach ensures that the model retains its original strengths while acquiring a profound understanding of German, **setting a new benchmark in bilingual language model proficiency.** # Table of Contents 1. [Overview of all Her0 models](#all-hero-models) 2. [Model Details](#model-details) - [Prompt template](#prompt-template) - [Training Dataset](#training-dataset) - [Merge Procedure](#merge-procedure) 3. [Evaluation](#evaluation) - [GPT4ALL](#gpt4all) - [Language Model evaluation Harness](#language-model-evaluation-harness) - [BigBench](#bbh) - [MMLU](#mmlu) - [TruthfulQA](#truthfulqa) - [MT-Bench (German)](#mt-bench-german) - [MT-Bench (English)](#mt-bench-english) - [Additional German Benchmark results](#additional-german-benchmark-results) 5. [Disclaimer](#disclaimer) 6. [Contact](#contact) 7. [Collaborations](#collaborations) 8. [Acknowledgement](#acknowledgement) ## All HerO Models | Model | HF | GPTQ | GGUF | AWQ | |-------|-------|-------|-------|-------| | SauerkrautLM-7b-HerO | [Link](https://huggingface.co/VAGOsolutions/SauerkrautLM-7b-HerO) | coming soon | coming soon | coming soon | ## Model Details **SauerkrautLM-7b-HerO** - **Model Type:** SauerkrautLM-7b-HerO is an auto-regressive language model based on the transformer architecture - **Language(s):** English, German - **License:** APACHE 2.0 - **Contact:** [Website](https://vago-solutions.de/#Kontakt) [David Golchinfar](mailto:golchinfar@vago-solutions.de) ### Training Dataset: SauerkrautLM-7b-HerO was trained with mix of German data augmentation and translated data. We found, that only a simple translation of training data can lead to unnatural German phrasings. Data augmentation techniques were used to grant grammatical, syntactical correctness and a more natural German wording in our training data. ### Merge Procedure: SauerkrautLM-7b-HerO was merged on 1 A100 with [mergekit](https://github.com/cg123/mergekit). The merged model contains [OpenHermes-2.5-Mistral-7B](https://huggingface.co/teknium/OpenHermes-2.5-Mistral-7B) and [Open-Orca/Mistral-7B-OpenOrca](https://huggingface.co/Open-Orca/Mistral-7B-OpenOrca). We applied the gradient SLURP method. ### Prompt Template: ``` <|im_start|>system Du bist Sauerkraut-HerO, ein großes Sprachmodell, das höflich und kompetent antwortet. Schreibe deine Gedanken Schritt für Schritt auf, um Probleme sinnvoll zu lösen. <|im_end|> <|im_start|>user Wie geht es dir?<|im_end|> <|im_start|>assistant Mir geht es gut!<|im_end|> <|im_start|>user Bitte erkläre mir, wie die Zusammenführung von Modellen durch bestehende Spitzenmodelle profitieren kann.<|im_end|> <|im_start|>assistant ``` ## Evaluation ### GPT4ALL: *Compared to relevant German Closed and Open Source models* ![GPT4ALL diagram](https://vago-solutions.de/wp-content/uploads/2023/11/GPT4All.png "SauerkrautLM-7b-HerO GPT4ALL Diagram") ![GPT4ALL table](https://vago-solutions.de/wp-content/uploads/2023/11/GPT4All-Tabelle.png "SauerkrautLM-7b-HerO GPT4ALL Table") ### Language Model evaluation Harness: *Compared to Aleph Alpha Luminous Models* ![Harness](https://vago-solutions.de/wp-content/uploads/2023/11/Luminous-comparison.png "SauerkrautLM-7b-HerO Harness") **performed with newest Language Model Evaluation Harness* ### BBH: ![BBH](https://vago-solutions.de/wp-content/uploads/2023/11/BigBench.png "SauerkrautLM-7b-HerO BBH") **performed with newest Language Model Evaluation Harness* ### MMLU: *Compared to Big Boy LLMs (Grok0,Grok1,GPT3.5,GPT4)* ![MMLU](https://vago-solutions.de/wp-content/uploads/2023/11/MMLU-Benchmark.png "SauerkrautLM-7b-HerO MMLU") ### TruthfulQA: *Compared to OpenAI Models (GPT3.5,GPT4)* ![TruthfulQA](https://vago-solutions.de/wp-content/uploads/2023/11/Truthfulqa-Benchmark.png "SauerkrautLM-7b-HerO TruthfulQA") ### MT-Bench (German): ![MT-Bench German Diagram](https://vago-solutions.de/wp-content/uploads/2023/11/MT-Bench-German.png "SauerkrautLM-7b-HerO MT-Bench German Diagram") ``` ########## First turn ########## score model turn SauerkrautLM-70b-v1 1 7.25000 SauerkrautLM-7b-HerO <--- 1 6.96875 SauerkrautLM-7b-v1-mistral 1 6.30625 leo-hessianai-13b-chat 1 6.18750 SauerkrautLM-13b-v1 1 6.16250 leo-mistral-hessianai-7b-chat 1 6.15625 Llama-2-70b-chat-hf 1 6.03750 vicuna-13b-v1.5 1 5.80000 SauerkrautLM-7b-v1 1 5.65000 leo-hessianai-7b-chat 1 5.52500 vicuna-7b-v1.5 1 5.42500 Mistral-7B-v0.1 1 5.37500 SauerkrautLM-3b-v1 1 3.17500 Llama-2-7b 1 1.28750 open_llama_3b_v2 1 1.68750 ########## Second turn ########## score model turn SauerkrautLM-70b-v1 2 6.83125 SauerkrautLM-7b-HerO <--- 2 6.30625 vicuna-13b-v1.5 2 5.63125 SauerkrautLM-13b-v1 2 5.34375 SauerkrautLM-7b-v1-mistral 2 5.26250 leo-mistral-hessianai-7b-chat 2 4.99375 SauerkrautLM-7b-v1 2 4.73750 leo-hessianai-13b-chat 2 4.71250 vicuna-7b-v1.5 2 4.67500 Llama-2-70b-chat-hf 2 4.66250 Mistral-7B-v0.1 2 4.53750 leo-hessianai-7b-chat 2 2.65000 SauerkrautLM-3b-v1 2 1.98750 open_llama_3b_v2 2 1.22500 Llama-2-7b 2 1.07500 ########## Average ########## score model SauerkrautLM-70b-v1 7.040625 SauerkrautLM-7b-HerO <--- 6.637500 SauerkrautLM-7b-v1-mistral 5.784375 SauerkrautLM-13b-v1 5.753125 vicuna-13b-v1.5 5.715625 leo-mistral-hessianai-7b-chat 5.575000 leo-hessianai-13b-chat 5.450000 Llama-2-70b-chat-hf 5.350000 SauerkrautLM-v1-7b 5.193750 vicuna-7b-v1.5 5.050000 Mistral-7B-v0.1 4.956250 leo-hessianai-7b-chat 4.087500 SauerkrautLM-3b-v1 2.581250 open_llama_3b_v2 1.456250 Llama-2-7b 1.181250 ``` **performed with the newest FastChat Version* ### MT-Bench (English): ![MT-Bench English Diagram](https://vago-solutions.de/wp-content/uploads/2023/11/MT-Bench-Englisch.png "SauerkrautLM-7b-HerO MT-Bench English Diagram") ``` ########## First turn ########## score model turn OpenHermes-2.5-Mistral-7B 1 8.21875 SauerkrautLM-7b-HerO <--- 1 8.03125 Mistral-7B-OpenOrca 1 7.65625 neural-chat-7b-v3-1 1 7.22500 ########## Second turn ########## score model turn OpenHermes-2.5-Mistral-7B 2 7.1000 SauerkrautLM-7b-HerO <--- 2 6.7875 neural-chat-7b-v3-1 2 6.4000 Mistral-7B-OpenOrca 2 6.1750 ########## Average ########## score model OpenHermes-2.5-Mistral-7B 7.659375 SauerkrautLM-7b-HerO <--- 7.409375 Mistral-7B-OpenOrca 6.915625 neural-chat-7b-v3-1 6.812500 ``` **performed with the newest FastChat Version* ### Additional German Benchmark results: ![GermanBenchmarks](https://vago-solutions.de/wp-content/uploads/2023/11/German-benchmarks.png "SauerkrautLM-7b-HerO German Benchmarks") *performed with newest Language Model Evaluation Harness ## Disclaimer We must inform users that despite our best efforts in data cleansing, the possibility of uncensored content slipping through cannot be entirely ruled out. 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. 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. These models may be employed for commercial purposes, and the Apache 2.0 remains applicable and is included with the model files.   ## Contact If you are interested in customized LLMs for business applications, please get in contact with us via our website or contact us at [Dr. Daryoush Vaziri](mailto:vaziri@vago-solutions.de). We are also grateful for your feedback and suggestions.   ## Collaborations We are also keenly seeking support and investment for our startup, VAGO solutions, 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. ## Acknowledgement Many thanks to [OpenOrca](https://huggingface.co/Open-Orca) and [teknium](https://huggingface.co/teknium) for providing such valuable models to the Open-Source community. [Built with Axolotl](https://github.com/OpenAccess-AI-Collective/axolotl)