List your company

Question Answering on TriviaQA (val)

74.8FT F1FLASH-Quad
66.89668.9487173.052Feb 21, 2022

Evaluation results

MethodLinks
FLASH-QuadContext Length=4096, Model Size=~110M, Hardware=32 TPU-v4 cores, Head size (s)=5122022.0274.81.67
Transformer+Context Length=4096, Model Size=~110M, Hardware=32 TPU-v4 cores2022.0274.21
FLASHContext Length=4096, Model Size=~110M, Hardware=32 TPU-v4 cores, Head size (s)=512, Attention mechanism (first-to-all)=true2022.0273.92.69
FLASHContext Length=4096, Model Size=~110M, Hardware=32 TPU-v4 cores, Head size (s)=5122022.0273.32.6
FLASH-QuadContext Length=4096, Model Size=~110M, Hardware=32 TPU-v4 cores, Head size (s)=1282022.0272.71.79
CombinerContext Length=4096, Model Size=~110M, Hardware=32 TPU-v4 cores2022.0267.22.75