Update README.md

README.md

@@ -8,14 +8,30 @@ tags:
 license: apache-2.0
 language:
 - en
+pipeline_tag: text-generation
+library_name: transformers
 ---
 
-# Uploaded finetuned  model
+This is a content fidelity model. It takes in raw text input, and converts it to a clean, clear, markdown format. Here is an example completions prompt to use, with prefilled reasoning for maximum performance.
 
-- **Developed by:** textcleanlm
-- **License:** apache-2.0
-- **Finetuned from model :** unsloth/gpt-oss-20b-BF16
+Replace `developer_message` with anything you want. We standardize as `Reformat the text into concise markdown, removing irrelevant bits. Strength: high`, but you can tweak the strength, change the prompt, etc.
 
-This gpt_oss model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
+`user_input` is the raw content that you want to extract from.
 
-[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
+Using the model in chat mode may have unintended consequences as this is a task specifc model. YMMV
+
+```
+<|start|>system<|message|>You are ChatGPT, a large language model trained by OpenAI.
+Knowledge cutoff: 2024-06
+Current date: 2025-08-13
+
+Reasoning: medium
+
+# Valid channels: analysis, commentary, final. Channel must be included for every message.
+Calls to these tools must go to the commentary channel: 'functions'.<|end|><|start|>developer<|message|>
+{developer_message}<|end|><|start|>user<|message|>
+
+{user_input}
+
+<|end|><|start|>assistant<|channel|>analysis<|message|>The user wants reformatted text into concise markdown, removing irrelevant bits, high strength. Let's do it concise.<|end|><|start|>assistant<|channel|>final<|message|>
+```