Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Training Efficiency on Large Language Model Pre-training

46.2Model FLOPs Utilization

PaLM

20.30427.02733.7540.473Apr 5, 2022
Updated 1mo ago

Evaluation Results

MethodLinks
2022.04
46.257.8
2022.04
32.5-
2022.04
30.2-
2022.04
21.3-