This is quantized 4 bit, 5 bit, 8 bit and fp16 model versions of the bling-phi-2-v0 by llmware. Quantized models are in gguf format and done using llama.cpp

For more info about prompt template refer the original model repo.

Original Model Details:

Creator: llmware

Link: https://huggingface.co/llmware/bling-phi-2-v0