mradermacher/CodeShell-7B-GGUF
Same weights, converted vs WisdomShell/CodeShell-7B. Same weights stored in another dtype (BF16 → F16/F32), plus 1 new tensors.
Tensors changed: Attention 0/168, Norms 0/170, MLP 0/168, Embeddings 0/1
Same weights, converted vs WisdomShell/CodeShell-7B. Same weights stored in another dtype (BF16 → F16/F32), plus 1 new tensors.
Tensors changed: Attention 0/168, Norms 0/170, MLP 0/168, Embeddings 0/1