Update README.md

README.md

@@ -7,6 +7,13 @@ license: llama3.1
 language:
 - en
 pipeline_tag: text-generation
+base_model:
+- OpenBuddy/openbuddy-llama3.1-8b-v22.2-131k
+- THUDM/LongWriter-llama3.1-8b
+- akjindal53244/Llama-3.1-Storm-8B
+- aifeifei798/DarkIdol-Llama-3.1-8B-Instruct-1.2-Uncensored
+- ValiantLabs/Llama3.1-8B-Enigma
+- agentlans/Llama3.1-vodka
 ---
 # Llama3.1-Dark-Enigma
 
@@ -22,27 +29,25 @@ This model can be used in various applications where natural language processing
 
 ## Data Overview
 The model is built by merging several Llama 3.1 8B text models selected for their diverse layer weights. This fusion aims to leverage the strengths of each component, resulting in a more robust and versatile AI tool.
+- [OpenBuddy/openbuddy-llama3.1-8b-v22.2-131k](https://huggingface.co/OpenBuddy/openbuddy-llama3.1-8b-v22.2-131k)
+- [THUDM/LongWriter-llama3.1-8b](https://huggingface.co/THUDM/LongWriter-llama3.1-8b)
+- [akjindal53244/Llama-3.1-Storm-8B](https://huggingface.co/akjindal53244/Llama-3.1-Storm-8B)
+- [aifeifei798/DarkIdol-Llama-3.1-8B-Instruct-1.2-Uncensored](https://huggingface.co/aifeifei798/DarkIdol-Llama-3.1-8B-Instruct-1.2-Uncensored)
+- [ValiantLabs/Llama3.1-8B-Enigma](https://huggingface.co/ValiantLabs/Llama3.1-8B-Enigma)
 
-- openbuddy-llama3.1-8b-v22.2-131k
-- LongWriter-llama3.1-8b
-- Llama-3.1-Storm-8B
-- DarkIdol-Llama-3.1-8B-Instruct-1.2-Uncensored
-- Llama3.1-8B-Enigma
-
-Those models were merged onto Llama3.1-vodka using mergekit's `model\_stock` method with equal weights.
+Those models were merged onto [agentlans/Llama3.1-vodka](https://huggingface.co/agentlans/Llama3.1-vodka) using [mergekit](https://github.com/arcee-ai/mergekit)'s `model_stock` method where each model is equally weighted.
 
 ## Performance Evaluation
 While specific performance metrics are not provided, users can expect high-quality output when using effective prompting techniques and grounded input texts. The model's uncensored nature ensures it doesn't shy away from complex or sensitive topics.
 
-In fact, most of this dataset card was generated by the model itself.
+In fact, most of this model card was generated by the model itself.
 
 ## Limitations
 Users should note the following:
-- Do not rely solely on the model's output; always validate its results.
+- Do not rely solely on the model's output. Always validate its results.
   - As a 8B parameter model, it does poorly on closed book factual questions and answers.
 - For optimal performance, use good prompting strategies to guide the model effectively.
 - Be cautious when processing text that may contain biases or inaccuracies.
-- The quality may vary for languages other than English.
 - The model can't connect to the Internet and it doesn't know how to use specific APIs, libraries, or frameworks.
 
 ## Bias and Fairness Analysis