DevQuasar-6/Orion-zhen.phi-4-abliterated-GGUF
Same weights, converted vs Orion-zhen/phi-4-abliterated. Same weights stored in another dtype (BF16/F16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/81, MLP 0/80, Attention 0/80, lm_head 0/1
Same weights, converted vs Orion-zhen/phi-4-abliterated. Same weights stored in another dtype (BF16/F16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/81, MLP 0/80, Attention 0/80, lm_head 0/1