Update README.md

README.md

@@ -15,11 +15,11 @@ datasets:
 - philschmid/test_german_squad
 ---
 # Introduction
-This repository contains the model [jphme/Llama-2-13b-chat-german](https://huggingface.co/jphme/Llama-2-13b-chat-german) in GGUF format, for fast and easy inference with llama.cpp and similar LLM inference tools.
-This model was created and trained by [jphme](https://huggingface.co/jphme). It is a fine-tuned variant of Meta's [Llama2 13b Chat](https://huggingface.co/meta-llama/Llama-2-13b-chat) with a compilation of multiple instruction datasets in German language. 
+This repository contains the model [jphme/Llama-2-13b-chat-german](https://huggingface.co/jphme/Llama-2-13b-chat-german) in GGUF format.
+This model was created by [jphme](https://huggingface.co/jphme). It's a fine-tuned variant of Meta's [Llama2 13b Chat](https://huggingface.co/meta-llama/Llama-2-13b-chat) with a compilation of multiple instruction datasets in German language. 
 
-## AIML Profile
-The Model Profile (config.aiml file) describes relevant properties, configuration and rules for the AI model in a standardized, digestible and easy-to-read way.
+## Model Profile
+The AIML Profile stored in a file named "config.aiml" contains all relevant configuration parameters, properties and rules for securely deploying the AI model without hassle.
 
 ### General Information
 |Attribute|Details|
@@ -33,10 +33,15 @@ The Model Profile (config.aiml file) describes relevant properties, configuratio
 |Attribute|Details|
 |----------------------------|--------------------------------------------------------------------------------------------------------------|
 | **Type**                   | Large Language Model                                                                                         |
-| **Function**               | Text Generation                                                                                              |
+| **Pipeline**               | Text Generation                                                                                              |
 | **Architecture**           | Transformers                                                                                                 |
-| **Variables**              | {"llm_flavor":"llama", "llm_prompt_template":"llama2", "devices":"gpu[0,1,2,3],cpu[0]", "key":"value"}       |
+| **Variables**              | {"llm_languages":"en,de,nl,it,fr", 
+                                "llm_flavor":"llama",
+                                "llm_prompt_template":"llama2",
+                                "devices":"gpu[0,1,2,3],cpu[0]",
+                                "key":"value"}                                                                                              |
 | **Filetype**               | GGUF                                                                                                         |
+| **InferenceTools**         | Llama.cpp, Text Generation Inference (TGI), h2oGPT Server, KoboldCpp, Custom                                 |
 | **Compression**            | 8 Bit, 5 Bit (K_M), 4 Bit (K_M)                                                                              |
 | **CompressionMethod**      | llama.cpp - convert.py Script                                                                                |
 | **Notes**                  | First, a FP16 GGUF file was generated, and then quantized it to 8, 4 (K_M) and 5 (K_M) Bit with llama.cpp/quantize |                                                                                                       |
@@ -44,7 +49,7 @@ The Model Profile (config.aiml file) describes relevant properties, configuratio
 ### Customization
 |Attribute|Details|
 |----------------------------|-----------------------------------------------------------------------------------------------------------------|
-| **Type**                   | finetune_full (e.g. none, finetune_lora, finetune_qlora, finetune_full)                                          |
+| **Type**                   | finetune_full (e.g. none, finetune_lora, finetune_qlora, finetune_full)                                         |
 | **Class**                  | Instruct, Chat |
 | **Datasets**               | {"[Prorietary German Conversation Dataset](https://placeholder.ocal/dataset)", "[German & German legal SQuAD](https://placeholder.local/dataset)" |
 | **Notes**                  | The datasets were augmented with rows containing "wrong" contexts, in order to improve factual RAG performance. |