Curious to try this with diloco, would still do bs=1 on the inner optimizer and still get benefits of data parallelism
Micah Goldblum
Micah GoldblumJul 10, 22:12
🚨 Did you know that small-batch vanilla SGD without momentum (i.e. the first optimizer you learn about in intro ML) is virtually as fast as AdamW for LLM pretraining on a per-FLOP basis? 📜 1/n
1.89K