DeepSpeed
DeepSpeed copied to clipboard
Fix redundant shared_params in zero_to_fp32.py
state_dict["module"] has redundant params that were mistakenly recorded in shared_params
Related: (1) https://github.com/microsoft/DeepSpeed/issues/3291 (2) https://github.com/microsoft/DeepSpeed/pull/3295
lets merge this?
@mayank31398, does this work in your testing?