Terracotta

Terracotta is a cutting-edge platform designed to enhance the workflow for developers and researchers working with large language models (LLMs). This intuitive and user-friendly platform allows you to manage, iterate, and evaluate your fine-tuned models with ease. With Terracotta, you can securely upload data, fine-tune models for various tasks like classification and text generation, and create comprehensive evaluations to compare model performance using both qualitative and quantitative metrics. Our tool supports connections to major providers like OpenAI and Cohere, ensuring you have access to a broad range of LLM capabilities. Terracotta is the creation of Beri Kohen and Lucas Pauker, AI enthusiasts and Stanford graduates, who are dedicated to advancing LLM development. Join our email list to stay informed on the latest updates and features that Terracotta has to offer.

Key Features

• Manage Many Models: Centrally handle all your fine-tuned models in one convenient place.

• Iterate Quickly: Streamline the process of model improvement with fast qualitative and quantitative evaluations.

• Multiple Providers: Seamlessly integrate with services from OpenAI and Cohere to supercharge your development process.

• Upload Your Data: Upload and securely store your datasets for the fine-tuning of models.

• Create Evaluations: Conduct in-depth comparative assessments of model performances leveraging metrics like accuracy BLEU and confusion matrices.

FAQ

What is Terracotta?

Terracotta is a platform that enables rapid and intuitive experimentation with large language models, facilitating model management, iteration, and evaluation.

How do I sign in to Terracotta?

You can sign in using your Google or GitHub accounts to start working with Terracotta.

What is the first step in using Terracotta for fine-tuning models?

By uploading your data to Terracotta, you can securely store it and later use it for fine-tuning your models.

What types of model fine-tuning does Terracotta support?

Terracotta allows you to fine-tune models on your uploaded data for both classification and text generation tasks.

How does qualitative and quantitative evaluation work on Terracotta?

Qualitative evaluation involves feeding prompts to various models and comparing their outputs, while quantitative evaluation includes the use of different metrics like accuracy, BLEU scores, and confusion matrices.