build-small-hackathon/Nemotron-nano-4b-escape-room

Same weights, converted vs unsloth/NVIDIA-Nemotron-3-Nano-4B-GGUF. Same weights stored in another dtype (BF16/F32 → BF16), plus 21 new tensors.

Tensors changed: Embeddings 0/1, Norms 0/43, lm_head 0/1

On the Hub · Downloads on Model Pulse