nvidia/NVIDIA-NemotronLabs-AI-for-Media-Sports-Tennis

Partial fine-tune vs nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16. Partial change: 84% of the weights changed (attention, lm_head, mlp, norms, other), in 52 of 52 layers.

Tensors changed: Norms 117/565, Embeddings 0/4, other 165/294, MLP 1579/1812, Attention 24/320, lm_head 1/1

1 derivatives checked