Sunday, May 01, 2022

Google: Pathways Language Model (PaLM): Scaling to 540 Billion Parameters for Breakthrough Performance

Here is the latest and greatest in computational language modeling! Very impressive!

Several well known researchers authored this paper: Jeff Dean, Orhan Firat, Jacob Devlin (of BERT fame), Noam Shazeer, Barret Zoph.

However, here again when you take a look at their research paper you will notice the intrusion of the latest ideologies and virtue signalling into their research like gender and other exaggerated "societal biases" or language toxicity or minority languages or carbon neutral etc..

"... Last year Google Research announced our vision for Pathways, a single model that could generalize across domains and tasks while being highly efficient. An important milestone toward realizing this vision was to develop the new Pathways system to orchestrate distributed computation for accelerators. In “PaLM: Scaling Language Modeling with Pathways”, we introduce the Pathways Language Model (PaLM), a 540-billion parameter, dense decoder-only Transformer model trained with the Pathways system, which enabled us to efficiently train a single model across multiple TPU v4 Pods. We evaluated PaLM on hundreds of language understanding and generation tasks, and found that it achieves state-of-the-art few-shot performance across most tasks, by significant margins in many cases. ...
PaLM demonstrates the first large-scale use of the Pathways system to scale training to 6144 chips, the largest TPU-based system configuration used for training to date. The training is scaled using data parallelism at the Pod level across two Cloud TPU v4 Pods, while using standard data and model parallelism within each Pod. This is a significant increase in scale compared to most previous LLMs, which were either trained on a single TPU v3 Pod (e.g., GLaM, LaMDA), used pipeline parallelism to scale to 2240 A100 GPUs across GPU clusters (Megatron-Turing NLG) or used multiple TPU v3 Pods (Gopher) with a maximum scale of 4096 TPU v3 chips. ...
PaLM shows breakthrough capabilities on numerous very difficult tasks. ..."

Google AI Blog: Pathways Language Model (PaLM): Scaling to 540 Billion Parameters for Breakthrough Performance

PaLM: Scaling Language Modeling with Pathways (open access; I am still reading this paper)

No comments: