Update README.md

README.md

@@ -25,6 +25,25 @@ Coupled with the German Sauerkraut dataset, which consists of a mix of augmented
 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)
+    - [MT-Bench (German)](#mt-bench-(german))
+    - [MT-Bench (English)](#mt-bench-(english))
+    - [Language Model evaluation Harness](#language-model-evaluation-harness)
+    - [BigBench](#BBH)
+    - [GPT4ALL](#gpt4all)
+    - [Additional German Benchmark results](#additional-german-benchmark-results)
+    - [GPT4ALL](#gpt4all)
+4. [Disclaimer](#disclaimer)
+5. [Contact](#contact)
+7. [Collaborations](#collaborations)
+8. [Acknowledgement](#acknowledgement)
+
 
 ## All HerO Models
 
@@ -34,26 +53,26 @@ Our approach ensures that the model retains its original strengths while acquiri
 
 ## 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:[email protected])
 
-**Training Dataset:**
+#**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:**
+#**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.
 
 
-- **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:[email protected])
 
-**Prompt Template:**
+# **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.
@@ -67,7 +86,7 @@ Bitte erkläre mir, wie die Zusammenführung von Modellen durch bestehende Spitz
 <|im_start|>assistant
 ```
 ## Evaluation
-**MT-Bench (German)**
+#**MT-Bench (German)**
 ```
 ########## First turn ########## 
                                                            score 
@@ -126,7 +145,7 @@ SauerkrautLM-3b-v1                                  2.581250
 open_llama_3b_v2                                    1.456250 
 Llama-2-7b                                          1.181250 
 ```
-**MT-Bench (English)**
+#**MT-Bench (English)**
 ```
 ########## First turn ########## 
                                                            score 
@@ -154,20 +173,20 @@ neural-chat-7b-v3-1                                 6.812500
 ```
 
 
-**Language Model evaluation Harness**
+#**Language Model evaluation Harness**
 Compared to Aleph Alpha Luminous Models:
 ![Harness](images/luminouscompare.PNG "SauerkrautLM-7b-HerO Harness")
 
 *performed with newest Language Model Evaluation Harness
-**BBH**
+#**BBH**
 ![BBH](images/bbh.PNG "SauerkrautLM-7b-HerO BBH")
 *performed with newest Language Model Evaluation Harness
-**GPT4ALL**
+#**GPT4ALL**
 Compared to Aleph Alpha Luminous Models, LeoLM and EM_German:
 ![GPT4ALL diagram](images/gpt4alldiagram.PNG "SauerkrautLM-7b-HerO GPT4ALL Diagram")
 
 ![GPT4ALL table](images/gpt4alltable.PNG "SauerkrautLM-7b-HerO GPT4ALL Table")
-**Additional German Benchmark results**
+#**Additional German Benchmark results**
 ![GermanBenchmarks](images/germanbench.PNG "SauerkrautLM-7b-HerO German Benchmarks")
 *performed with newest Language Model Evaluation Harness
 ## Disclaimer