mradermacher/Qwen3.8-4B-Distill-GGUF
Same weights, converted vs empero-ai/Qwen3.8-4B-Distill. Same weights stored in another dtype (BF16 → F16/F32), plus 39 new tensors.
Tensors changed: Embeddings 0/1, Norms 0/81, MLP 0/96, Attention 0/32
Same weights, converted vs empero-ai/Qwen3.8-4B-Distill. Same weights stored in another dtype (BF16 → F16/F32), plus 39 new tensors.
Tensors changed: Embeddings 0/1, Norms 0/81, MLP 0/96, Attention 0/32