Vijay Anand Korthikanti
Vijay Anand Korthikanti
LGTM. @jaredcasper can you please take a final look?
@jiemingz @mathemakitten @sidsingh-nvidia @lmcafee-nvidia Can you please take a look at this MR?
@jiemingz @mathemakitten @sidsingh-nvidia can you please take a look at this MR?
@sidsingh-nvidia @lmcafee-nvidia @santhnm2 can you please take a look at this MR?
@lmcafee-nvidia @mathemakitten can you please sign off on this MR?
Did you try out storing the logits in bf16? It could save lot of memory. Not sure if we need this fusion.
Can we make this feature optional? Also, can we move the kernels to TE and use them in the Megatron core?
@deepakn94 can you please take a look at this MR?
@santhnm2 @tdene can you please take a look at this MR?
@fanshiqing can you please take a look at this MR?