Proctora

🚀 Proctora - A MoE Text Generation Model

Proctora is a Mixture of Experts (MoE) model designed for text generation. It combines different expert models to offer diverse capabilities, suitable for both role - playing (RP) tasks and providing factual answers.

🚀 Quick Start

Proctora is readily accessible for use in text - generation scenarios. You can start leveraging its power with appropriate text - generation tools.

✨ Features

• Powerful Base Model: Built upon the OpenPipe/mistral - ft - optimized - 1227 as the base, ensuring a solid foundation for text generation.
• Expert - Driven Design:
  • The first expert, SanjiWatsuki/Kunoichi - 7B, is specialized in RP tasks, enabling high - quality role - playing interactions.
  • The second expert, samir - fama/SamirGPT - v1, is dedicated to providing factual answers.
• Long Context Length: Thanks to the Mixtral architecture, Proctora has a natural context length of 32K, which is highly beneficial for handling long - form text.
• Good Performance: It scores 71.88 on the Openllm leaderboard, although the leaderboard score doesn't fully represent its true capabilities.
• Excellent in RP: In role - playing scenarios, it shows remarkable logical quality. It can better handle special power limitations, curse activations, and weaknesses compared to other models. It can also play in - character and make intelligent decisions when acting as a player.
• Uncensored: With the SanjiWatsuki/Kunoichi - 7B expert, the model is uncensored.

📄 License

This project is licensed under the CC - BY - NC - 4.0 license.

Proctora is a MoE model made of:

• OpenPipe/mistral - ft - optimized - 1227 as a base model
• SanjiWatsuki/Kunoichi - 7B as a first expert dedicated to RP tasks.
• samir - fama/SamirGPT - v1 as a second expert for factual answers.

Being based on Mixtral architecture it has a natural context length of 32K, which is great.

On Openllm leaderboard it achieves a score of 71.88 which is interesting to some extent but does not really reflect the intended capacities of the model.

This model has been originally produced as a result of experimentations with mergekit. Then among my collection of LLMs, Proctora has been selected to be the "grader" in an AI - RPG evaluation suite that I am currently building. Indeed, it produced the intended grades according to given rubrics more often than other "higher performing" models in the leaderboard.

However, I also tested it in various RP scenarii using text - generation - webui (putting the character card in the system parameters and/or other world information), and I was quite impressed by the quality of the logic (relatively to other popular RP models). For example, it took in account special powers limitations better than other models. Or it managed curse activations and weaknesses better than other models that are about twice the size. Also when acting as the player (and the user being the game master), Proctora was not only able to play in character but also sometimes to make clever decision to achieve its objectives.

Having the excellent SanjiWatsuki/Kunoichi - 7B as an expert, the model is uncensored. Use with caution.

⚠️ Important Note

The model is uncensored. Use it with caution.

Support Me Here!

My Blog