Qwen/Qwen-72B-Chat

Full fine-tune vs Qwen/Qwen-72B. Fine-tune: 100% of the weights changed, across 80 of 80 layers.

Tensors changed: Embeddings 1/1, Attention 240/240, Norms 160/161, MLP 240/240, lm_head 1/1

2 derivatives checked

1 other repos hold byte-for-byte identical weight files.