maldv/Shisutemu-Masuta-Q3-32B

Full fine-tune vs DMindAI/DMind-1. Fine-tune: 98% of the weights changed, across 64 of 64 layers.

Tensors changed: Embeddings 0/1, Norms 222/257, MLP 192/192, Attention 256/256, lm_head 1/1

2 derivatives checked