Ahmadzei's picture
added 3 more tables for large emb model
5fa1a76
Simply call model.push_to_hub() after loading it in 4-bit precision.