mradermacher/Qwen1.5-0.4B-Chat-GGUF
Same weights, converted vs nathanrchn/Qwen1.5-0.4B-Chat. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/41, MLP 0/60, Attention 0/140
Same weights, converted vs nathanrchn/Qwen1.5-0.4B-Chat. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/41, MLP 0/60, Attention 0/140