mradermacher/Qwen2.5-2B-Instruct-GGUF
Same weights, converted vs win10/Qwen2.5-2B-Instruct. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/105, MLP 0/156, Attention 0/364
Same weights, converted vs win10/Qwen2.5-2B-Instruct. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/105, MLP 0/156, Attention 0/364