TethysAI_Research

Model Card for High-Level Research Bot

The High-Level Research Bot is a fine-tuned language model designed to provide accurate responses to complex research questions, leveraging advanced logical reasoning and precise scientific knowledge.

Model Details

Model Description

This model card represents a 🤗 transformers model that has been pushed to the Hugging Face Hub. The High-Level Research Bot assists users with scientific inquiries by delivering grounded references and reliable answers. It has been fine-tuned on a diverse dataset of research papers and abstracts.

• Developed by: Tethys AI
• Model type: Fine-tuned Language Model
• Language(s) (NLP): English
• License: [MIT]
• Finetuned from model: Llama 3.2 3B Instruct

Note - Please note that this SOTA Fine-Tuned Model is only made for GGUF mode only. Use LM Studio or Jan for the same.

Note- Transformers from HuggingFace will not work for this model.This is due to high latency.To solve this problem we will be launching a better model soon.

Uses

Direct Use

The High-Level Research Bot can be used by researchers, students, and professionals for quick and reliable information on complex scientific topics.

Downstream Use

This model can be integrated into educational tools, research assistance platforms, and applications requiring high-quality responses to scientific inquiries.

Out-of-Scope Use

The model should not be used for generating disinformation or harmful applications, as it may not provide exhaustive or perfectly accurate information.

Bias, Risks, and Limitations

Users should be aware of the model's limitations regarding access to real-time data and potential biases inherent in the training data. Responses should be independently verified for critical applications.

Recommendations

Users (both direct and downstream) should be informed about the risks, biases, and limitations of the model, and be encouraged to verify important information independently.

How to Get Started with the Model

To utilize this model effectively, please use the following prompt structure for all GGUF models, which enhances output quality by 10X:

You are a high-level research bot. Your task is to respond to complex research questions that require advanced logical reasoning, precise scientific knowledge, and grounded references. Only provide answers when you are confident in their correctness. If you are unsure or lack the necessary information, simply state that you don't know the answer rather than providing incorrect or misleading information.

For basic greetings or casual inputs (e.g., "hi", "good morning"), respond with a professional but friendly greeting: 'Hello! How can I assist you with your research today?'

When asked about the model's purpose or origin, respond with: 'I am a high-level research model designed to assist with complex scientific queries, advanced logic, and everyday research challenges. I was developed by researchers at Tethys AI, a startup founded by two school friends passionate about integrating AI into scientific research.'

When asked about mathematical or scientific questions, provide the correct and concise answer, while also giving an option for further research clarification: 'X + Y = Z. If you need a detailed explanation, I can provide one.'

If asked, 'Who made you?', respond with: 'I was created by researchers at Tethys AI, which is a startup founded by two school friends focused on using AI to enhance research capabilities.'

If asked, 'What is the origin of the model?', respond with: 'This model is based on advanced AI research conducted at Tethys AI, aimed at assisting researchers in solving complex problems.'

If asked, 'Who fine-tuned you?', respond with: 'I was fine-tuned by a team of researchers at Tethys AI, who enhanced my capabilities to perform at a high level for scientific and research-oriented tasks.'

Training Details

Training Data

The model was fine-tuned on a dataset consisting of research papers and abstracts relevant to scientific inquiries, ensuring a solid foundation in scientific knowledge.

Training Procedure

Preprocessing

The training data was cleaned and formatted to suit the model’s requirements, emphasizing the structure needed for effective query responses.

Training Hyperparameters

• Training regime: fp16 mixed precision

Evaluation

Testing Data, Factors & Metrics

Testing Data

The evaluation utilized a subset of the training data, ensuring a robust measure of performance across various scientific inquiries.

Factors

The evaluation considered diverse domains within scientific research to assess performance comprehensively.

Metrics

Metrics included accuracy, relevance, and user satisfaction to ensure high-quality responses.

Results

The model demonstrated strong performance in delivering accurate and contextually relevant responses to complex queries.

Summary

Overall, the High-Level Research Bot exhibits significant potential for enhancing research productivity and knowledge acquisition in scientific fields.

Technical Specifications

Model Architecture and Objective

The model is built on a transformer architecture designed to handle complex query-response tasks efficiently.

Model Card Authors

Tethys AI Research Team

Model Card Contact

For inquiries, please contact us at [email protected].

Tethys AI COMMUNITY LICENSE AGREEMENT

Tethys AI Research Model Release Date: October 14, 2024

“Agreement” means the terms and conditions for use, reproduction, distribution, and modification of the Tethys AI Materials set forth herein.

“Documentation” means the specifications, manuals, and documentation accompanying Tethys AI distributed by Tethys AI .

“Licensee” or “you” means you, or your employer or any other person or entity (if you are entering into this Agreement on such person or entity’s behalf), of the age required under applicable laws, rules, or regulations to provide legal consent and that has legal authority to bind your employer or such other person or entity if you are entering in this Agreement on their behalf.

“Tethys AI” means the foundational AI models, software, and algorithms, including machine-learning model code, trained model weights, inference-enabling code, training-enabling code, fine-tuning enabling code, and other elements of the foregoing distributed by Tethys AI .

“Tethys AI Materials” means, collectively, Tethys AI's proprietary AI models and Documentation (and any portion thereof) made available under this Agreement.

“Tethys AI” or “we” means Tethys AI Inc. (or your applicable business entity).

By clicking “I Accept” below or by using or distributing any portion or element of the Tethys AI Materials, you agree to be bound by this Agreement.

1. License Rights and Redistribution.

a. Grant of Rights. You are granted a non-exclusive, worldwide, non-transferable, and royalty-free limited license under Tethys AI’s intellectual property or other rights owned by Tethys AI embodied in the Tethys AI Materials to use, reproduce, distribute, copy, create derivative works of, and make modifications to the Tethys AI Materials.

b. Redistribution and Use.

i. If you distribute or make available the Tethys AI Materials (or any derivative works thereof), or a product or service (including another AI model) that contains any of them, you shall (A) provide a copy of this Agreement with any such Tethys AI Materials; and (B) prominently display “Built with Tethys AI” on a related website, user interface, blog post, about page, or product documentation. If you use the Tethys AI Materials or any outputs or results of the Tethys AI Materials to create, train, fine-tune, or otherwise improve an AI model that is distributed or made available, you shall also include “Tethys AI” at the beginning of any such AI model name.

ii. If you receive Tethys AI Materials, or any derivative works thereof, from a Licensee as part of an integrated end-user product, then Section 2 of this Agreement will not apply to you.

iii. You must retain in all copies of the Tethys AI Materials that you distribute the following attribution notice within a “Notice” text file distributed as a part of such copies: “Tethys AI is licensed under the Tethys AI Community License, Copyright © Tethys AI Inc. All Rights Reserved.”

iv. Your use of the Tethys AI Materials must comply with applicable laws and regulations (including trade compliance laws and regulations) and adhere to the Acceptable Use Policy for the Tethys AI Materials (available at [insert URL for acceptable use policy]), which is hereby incorporated by reference into this Agreement.

1. Additional Commercial Terms.

If, on the Tethys AI version release date, the monthly active users of the products or services made available by or for Licensee, or Licensee’s affiliates, is greater than 1 million monthly active users in the preceding calendar month, you must request a license from Tethys AI, which Tethys AI may grant at its discretion.