afrideva/phi-2-GGUF
Same weights, converted vs microsoft/phi-2. Same weights stored in another dtype (F16 → F16/F32), plus 64 new tensors.
Tensors changed: Embeddings 0/1, Norms 0/66, MLP 0/128, Attention 0/64, lm_head 0/2
Same weights, converted vs microsoft/phi-2. Same weights stored in another dtype (F16 → F16/F32), plus 64 new tensors.
Tensors changed: Embeddings 0/1, Norms 0/66, MLP 0/128, Attention 0/64, lm_head 0/2