mradermacher/Refact-1_6B-fim-GGUF
Same weights, converted vs refactai/Refact-1_6B-fim. Same weights stored in another dtype (BF16 → F16/F32), plus 160 new tensors.
Tensors changed: lm_head 0/1, Norms 0/65, Attention 0/32, MLP 0/32, Embeddings 0/1
Same weights, converted vs refactai/Refact-1_6B-fim. Same weights stored in another dtype (BF16 → F16/F32), plus 160 new tensors.
Tensors changed: lm_head 0/1, Norms 0/65, Attention 0/32, MLP 0/32, Embeddings 0/1