T5 Xl Lm Adapt

🚀 Google's T5 Version 1.1 - LM-Adapted

This is a version of Google's T5 model, which has been improved and pre - trained for better performance in natural language processing tasks.

🚀 Quick Start

This README provides detailed information about Google's T5 Version 1.1 - LM - Adapted, including its improvements, pre - training details, and related references.

✨ Features

T5 Version 1.1 - LM Adapted offers the following enhancements compared to the original T5 model:

• It uses GEGLU activation in the feed - forward hidden layer instead of ReLU. For more details, refer to here.
• Dropout was disabled during pre - training (resulting in a quality improvement), and it should be re - enabled during fine - tuning.
• It was pre - trained only on C4 without mixing in downstream tasks.
• There is no parameter sharing between the embedding and classifier layer.
• "xl" and "xxl" replace "3B" and "11B". The model shapes are slightly different, with a larger d_model and smaller num_heads and d_ff.

This model is pre - trained on both the denoising and language modeling objectives. Specifically, this checkpoint is initialized from T5 Version 1.1 - XL and then trained for an additional 100K steps on the LM objective discussed in the T5 paper. This adaptation enhances the model's ability for prompt tuning.

Note: A well - known fine - tuned version of the T5 Version 1.1 - LM Adapted model is BigScience's T0pp.

📚 Documentation

Pretraining Dataset

C4

Other Community Checkpoints

here

Paper

Exploring the Limits of Transfer Learning with a Unified Text - to - Text Transformer

Authors

Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, Peter J. Liu

Abstract

Transfer learning, where a model is first pre - trained on a data - rich task before being fine - tuned on a downstream task, has emerged as a powerful technique in natural language processing (NLP). The effectiveness of transfer learning has given rise to a diversity of approaches, methodology, and practice. In this paper, we explore the landscape of transfer learning techniques for NLP by introducing a unified framework that converts every language problem into a text - to - text format. Our systematic study compares pre - training objectives, architectures, unlabeled datasets, transfer approaches, and other factors on dozens of language understanding tasks. By combining the insights from our exploration with scale and our new “Colossal Clean Crawled Corpus”, we achieve state - of - the - art results on many benchmarks covering summarization, question answering, text classification, and more. To facilitate future work on transfer learning for NLP, we release our dataset, pre - trained models, and code.

📄 License

This project is licensed under the Apache 2.0 License.