 Matthias Reso
|
9def4fbafd
Remove micro_batch_training parameter and replace with gradient_accumulation_steps
|
2 jaren geleden |
 Brian Vaughan
|
3faf005226
fix a bug in the config for use_fast_kernels
|
2 jaren geleden |
 lchu
|
feaa344af3
resolve conflicts
|
2 jaren geleden |
 Hamid Shojanazeri
|
44ef280d31
adding flash attention and xformer memory efficient through PT SDPA
|
2 jaren geleden |
|
lchu
|
895dfcea30
add nightly check for using low_cpu_fsdp mode
|
2 jaren geleden |
 chauhang
|
4767f09ecd
Initial commit
|
2 jaren geleden |
