Closed ananyahjha93 closed 4 months ago
There is an order of magnitude difference between the losses between the two setups. @epwalsh can you sanity check the OLMo grad_clipping code for FSDP no_shard?
There is an order of magnitude difference between the losses between the two setups. @epwalsh can you sanity check the OLMo grad_clipping code for FSDP no_shard?