Open-source pre-training implementation of Google's LaMDA research paper in PyTorch. The totally not sentient AI. This repository will cover the 2B parameter implementation of the pre-training architecture as that is likely what most can afford to train. You can review Google's latest blog post from 2022 which details LaMDA here. You can also view their previous blog post from 2021 on the model.
Features
- Open-source pre-training implementation of Google's LaMDA research paper in PyTorch
- T5 Relative Positional Bias in Attention
- Gated GELU Activation in the Feed forward layer
- GPT-like Decoder Only architecture
- Autoregressive with Top-k sampling
- Sentencepiece Byte-pair encoded tokenizer
License
MIT LicenseFollow LaMDA-pytorch
You Might Also Like
Our Free Plans just got better! | Auth0 by Okta
You asked, we delivered! Auth0 is excited to expand our Free and Paid plans to include more options so you can focus on building, deploying, and scaling applications without having to worry about your secuirty. Auth0 now, thank yourself later.
Rate This Project
Login To Rate This Project
User Reviews
Be the first to post a review of LaMDA-pytorch!