mradermacher/Samantha-v7-4b-lesscensored-GGUF
Same weights, converted (fewer layers) vs Guilherme34/Samantha-v7-4b-lesscensored. mradermacher/Samantha-v7-4b-lesscensored-GGUF matches part of Guilherme34/Samantha-v7-4b-lesscensored (444 of its 883 tensors; the rest of Guilherme34/Samantha-v7-4b-lesscensored has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/205, MLP 0/102, Attention 0/136