README.md

---
license: apache-2.0
tags:
- mistral
- Uncensored
- text-generation-inference
- transformers
- unsloth
- trl
- roleplay
- conversational
- rp
datasets:
- N-Bot-Int/Iris-Uncensored-R1
- N-Bot-Int/Moshpit-Combined-R2-Uncensored
- N-Bot-Int/Mushed-Dataset-Uncensored
- N-Bot-Int/Muncher-R1-Uncensored
- unalignment/toxic-dpo-v0.2
language:
- en
base_model:
- unsloth/mistral-7b-instruct-v0.3-bnb-4bit
pipeline_tag: text-generation
library_name: transformers
metrics:
- character
new_version: N-Bot-Int/MistThena7B-V2
---
<a href="https://ibb.co/GvDjFcVp"><img src="https://raw.githubusercontent.com/Nexus-Network-Interactives/HuggingfacePage/refs/heads/main/MistThena7B.webp" alt="image" border="0"></a>
# Official Quants are Uploaded By Us
  - [MistThena7B GGUF](https://huggingface.co/N-Bot-Int/MistThena7B-GGUF)
# Wider Quant Supports are Uploaded By mradermacher!
  - *Thank you so much for the Help mradermacher!*
  - [mrardermarcher's GGUF & Weight support](https://huggingface.co/mradermacher/MistThena7B-GGUF)
  - [mrardermarcher's GGUF & Weight support(i1)](https://huggingface.co/mradermacher/MistThena7B-i1-GGUF)


# MistThena7B - A.
- MistThena7B is our brand New AI boasting with An Even Bigger **7B** and Ditching **Llama3.2** for **Mistral** for **lightweight Finetuning**
  And Fast Training and Output. MistThena7B is designed to Ditch its Outer-score and Prioritize Total Roleplaying, Trained with **5x More** Dataset
  Compared to What We used At **OpenElla3-Llama3.2B**, Making this New Model Even More Competitive **Against Hallucinations, and Even More Better
  Textual Generations And Uncensored Output**

- MistThena7B Model **A** Does not suffer the same Prompting issue with **OpenElla3-Llama3.2B**, however please use ChatML style Prompting For Better
  Experience, And Remember to be aware of bias with the training dataset used, The **AI** model is Under **Apache 2.0** however
  **WE ARE NOT RESPONSIBLE TO YOUR USAGE, PROMPTING, AND WAYS ABOUT HOW YOU USE THE MODEL. PLEASE BE GUIDED OWN ACCORDING/WILL**

- MistThena7B Model **A** Outperforms OpenElla Family Model, However please keep in mind the Parameter Difference. It Outperforms Testing Benchmarks
  In **Roleplaying and Engaging with RP or Generation of Prompts**, You are Free to release a Benchmark.

- MistThena7B contains more Fine-tuned Dataset so please Report any issues found through our email
  [nexus.networkinteractives@gmail.com](mailto:nexus.networkinteractives@gmail.com)
  about any overfitting, or improvements for the future Model **B**,
  Once again feel free to Modify the LORA to your likings, However please consider Adding this Page
  for credits and if you'll increase its **Dataset**, then please handle it with care and ethical considerations
  
- MistThena is
  - **Developed by:** N-Bot-Int
  - **License:** apache-2.0
  - **Finetuned from model :** unsloth/mistral-7b-instruct-v0.3-bnb-4bit
  - **Sequential Trained from Model:** N-Bot-Int/OpenElla3-Llama3.2A
  - **Dataset Combined Using:** Mosher-R1(Propietary Software)

- Comparison Metric Score
    ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6633a73004501e16e7896b86/194HxA6P0846242q7u_Y0.png)
    - Metrics Made By **ItsMeDevRoland**
      Which compares:
       - **Deepseek R1 3B GGUF**
       - **Dolphin 3B GGUF**
       - **Hermes 3b Llama GGUFF**
       - **OpenElla3-Llama3.2B GGUFF**
      Which are All Ranked with the Same Prompt, Same Temperature, Same Hardware(Google Colab),
      To Properly Showcase the differences and strength of the Models

    - **THIS MODEL EXCELLS IN LONGER PROMPT AND STAYING IN CHARACTER BUT LAGS BEHIND DEEPSEEK-R1**
 
  - # THERE ARE YET TO BE RELEASED METRIC SCORE FOR THIS MODEL, PLEASE REMAIN PATIENT WHILST **ItsMeDevRoland** Released an Updated Report
      

- # Notice
  - **For a Good Experience, Please use**
    - Low temperature 1.5, min_p = 0.1 and max_new_tokens = 128


- # Detail card:
  - Parameter
    - 7 Billion Parameters
    - (Please visit your GPU Vendor if you can Run 7B models)

  - Training
    - 200 steps
      - N-Bot-Int/Iris-Uncensored-R1
    - 100
      - N-Bot-Int/Iris-Uncensored-R1(Reinforcement Training)
    - 100 steps
      - M-Datasets
    - 60 steps(DPO)
      - Unalignment/Toxic-DPO

  - Finetuning tool:
   - Unsloth AI
     - This mistral model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
    [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
   - Fine-tuned Using:
    - Google Colab