Update README.md

README.md

@@ -16,6 +16,10 @@ datasets:
 - grimulkan/LimaRP-augmented
 - KaraKaraWitch/PIPPA-ShareGPT-formatted
 ---
+This is an ExLlamaV2 quantized model in 4.7bpw of [mpasila/Llama-3-Instruct-LiPPA-8B](https://huggingface.co/mpasila/Llama-3-Instruct-LiPPA-8B) using the default calibration dataset with 8192 context length.
+
+# Original Model card:
+
 This is a merge of [mpasila/Llama-3-LiPPA-LoRA-8B](https://huggingface.co/mpasila/Llama-3-LiPPA-LoRA-8B).
 
 LoRA trained in 4-bit with 8k context using [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct/) as the base model for 1 epoch.
@@ -24,6 +28,8 @@ Dataset used is [mpasila/LimaRP-PIPPA-Mix-8K-Context](https://huggingface.co/dat
 
 This has been trained on the instruct model and not the base model. The model trained with the base model using the same dataset is here: [mpasila/Llama-3-LiPPA-8B](https://huggingface.co/mpasila/Llama-3-LiPPA-8B)
 
+This also seems to work fairly well for chatting.
+
 ### Prompt format: Llama 3 Instruct
 
 Unsloth changed assistant to gpt and user to human.