Sahaj Agarwal

icon indicating an email [email protected]

icon company Microsoft icon location Experienced in full stack NLP and model optimization

Results 1 comments of Sahaj Agarwal

Error when run reshard_fsdp on opt-IML 30b for inference

@Charles-ux-bit Assuming 30B-IML checkpoint is same as 30B checkpoint in architecture, a hack that works is to copy "shard_metadata" from the later and then use reshard_fsdp script