Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Training Efficiency on Llama3-8x70B Coarse-grained
Loading...
41.6
MFU
MCore w/ Folding
18.72
24.66
30.6
36.54
Apr 21, 2025
MFU
Updated 1mo ago
Evaluation Results
Method
Method
Links
MFU
MCore w/ Folding
GPUs=256, Global batch...
2025.04
41.6
MCore
GPUs=256, Global batch...
2025.04
38.8
FSDP + EP
GPUs=256, Global batch...
2025.04
19.6
Feedback
Search any
task
Search any
task
