Shinrali/Qwen3.8-Flash-Next-NVIDIA-Hybrid-FP8-LMHead-Single-Spark

Partial fine-tune vs Shinrali/Qwen3.8-Flash-Next-NVIDIA-Hybrid-Single-Spark. Partial change: 2% of the weights changed (lm_head), in 0 of 48 layers.

Tensors changed: Embeddings 0/138, Attention 0/764, MLP 0/1786, Norms 0/302, other 0/8, lm_head 1/1

On the Hub · Downloads on Model Pulse