mradermacher/Falcon-H1R-7B-GGUF
Same weights, converted vs tiiuae/Falcon-H1R-7B. Same weights stored in another dtype (BF16 → F16/F32), plus 44 new tensors.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/89, Attention 0/176, MLP 0/132
Same weights, converted vs tiiuae/Falcon-H1R-7B. Same weights stored in another dtype (BF16 → F16/F32), plus 44 new tensors.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/89, Attention 0/176, MLP 0/132