Update README.md

README.md

@@ -27,21 +27,25 @@ It uses the Unified AI Description Format (.aidf), which is a novel format with
 |----------------------------|--------------------------------------------------------------------------------------------------------------|
 | **Model**                  | [jphme/Llama-2-13b-chat-german](https://huggingface.co/jphme/Llama-2-13b-chat-german)                        |
 | **Type**                   | Text Generation                                                                                              |
-| **Finetuned for**          | Chat                                                                                                         |                                                                                              |
+| **Parameters**             | 70B                                                                                                          |
+| **Fine-tuned**             | Chat, Instruct                                                                                               |                                                                                              |
 | **Architecture**           | Transformers                                                                                                 | 
 | **File Format**            | GGUF                                                                                                         |
 | **Quantization Types**     | 8 Bit <br>5 Bit (K_M)                                                                                        |
-| **Tools used**             | llama.cpp (Commit 9e20231)                                                                                   |
+| **Tools used**             | llama.cpp (Commit 9e20231) for quantization to 8, 5 and 4 bit                                                |
 | **Model Creator**          | [jphme](https://huggingface.co/jphme)                                                                        |
 | **Training Data**          | Prorietary German Conversation Dataset, German SQuAD, German legal SQuAD data, augmented with "wrong" contexts, to improve factual RAG. For details see original model link. |
 
-**Profile Metadata**
-*Profile Type:* 
-*Profile Version:* v1.0
-## Replication Steps
+**Metadata**
+*Profile Type:* General AI Profile
+*Profile Version:* AIDF v1.0
+
+## Replicate
 1. Clone and install llama.cpp *(Commit: 9e20231)*.
 ```
-# Install llama.cpp by cloning the repo and compiling it.
+# Install llama.cpp by cloning the repo from Github.
+# When cloned, then:
+cd llama.cpp && make
 ```
 2. Use the provided `convert.py` file to convert the original model to GGUF with FP16 precision.
 ```