legraphista/Phi-3-medium-4k-instruct-IMat-GGUF
Same weights, converted vs microsoft/Phi-3-medium-4k-instruct. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/81, MLP 0/80, Attention 0/80, lm_head 0/1
Same weights, converted vs microsoft/Phi-3-medium-4k-instruct. Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/81, MLP 0/80, Attention 0/80, lm_head 0/1