Update README.md

README.md

@@ -12,41 +12,20 @@ base_model:
 - ResplendentAI/Paradigm_7B
 - jeiku/Re-Host_Limarp_Mistral
 library_name: transformers
-tags:
-- mergekit
-- merge
-
+license: apache-2.0
+language:
+- en
 ---
-# Adult_Aura
-
-This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
-
-## Merge Details
-### Merge Method
+# Aura v3
 
-This model was merged using the [Model Stock](https://arxiv.org/abs/2403.19522) merge method using [ResplendentAI/Paradigm_7B](https://huggingface.co/ResplendentAI/Paradigm_7B) as a base.
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/626dfb8786671a29c715f8a9/V_DYIcPMJ5_ijanQW_ap2.png)
 
-### Models Merged
+Aura v3 is an improvement with a significantly more steerable writing style. Out of th box it will prefer poetic prose, but if instructed, it can adopt a more approachable style. This iteration has erotica, RP data and NSFW pairs to provide a more compliant mindset.
 
-The following models were included in the merge:
-* [ResplendentAI/Paradigm_7B](https://huggingface.co/ResplendentAI/Paradigm_7B) + [jeiku/selfbot_256_mistral](https://huggingface.co/jeiku/selfbot_256_mistral)
-* [ResplendentAI/Paradigm_7B](https://huggingface.co/ResplendentAI/Paradigm_7B) + [jeiku/Theory_of_Mind_Mistral](https://huggingface.co/jeiku/Theory_of_Mind_Mistral)
-* [ResplendentAI/Paradigm_7B](https://huggingface.co/ResplendentAI/Paradigm_7B) + [jeiku/Alpaca_NSFW_Shuffled_Mistral](https://huggingface.co/jeiku/Alpaca_NSFW_Shuffled_Mistral)
-* [ResplendentAI/Paradigm_7B](https://huggingface.co/ResplendentAI/Paradigm_7B) + [jeiku/Luna_LoRA_Mistral](https://huggingface.co/jeiku/Luna_LoRA_Mistral)
-* [ResplendentAI/Paradigm_7B](https://huggingface.co/ResplendentAI/Paradigm_7B) + [jeiku/Re-Host_Limarp_Mistral](https://huggingface.co/jeiku/Re-Host_Limarp_Mistral)
+I recommend keeping the temperature around 1.5 or lower with a Min P value of 0.05. This model can get carried away with prose at higher temperature. I will say though that the prose of this model is distinct from the GPT 3.5/4 variant, and lends an air of humanity to the outputs. I am aware that this model is overfit, but that was the point of the entire exercise.
 
-### Configuration
+If you have trouble getting the model to follow an asterisks/quote format, I recommend asterisks/plaintext instead. This model skews toward shorter outputs, so be prepared to lengthen your introduction and examples if you want longer outputs.
 
-The following YAML configuration was used to produce this model:
+This model responds best to ChatML for multiturn conversations.
 
-```yaml
-models:
-  - model: ResplendentAI/Paradigm_7B+jeiku/selfbot_256_mistral
-  - model: ResplendentAI/Paradigm_7B+jeiku/Theory_of_Mind_Mistral
-  - model: ResplendentAI/Paradigm_7B+jeiku/Alpaca_NSFW_Shuffled_Mistral
-  - model: ResplendentAI/Paradigm_7B+jeiku/Luna_LoRA_Mistral
-  - model: ResplendentAI/Paradigm_7B+jeiku/Re-Host_Limarp_Mistral
-merge_method: model_stock
-base_model: ResplendentAI/Paradigm_7B
-dtype: bfloat16
-```
+This model, like all other Mistral based models, is compatible with a Mistral compatible mmproj file for multimodal vision capabilities in KoboldCPP.