as I understand I need .bin file to to do:

    ./export-lora

    -m models/7B/ggml-model-f16.gguf

    -o models/7B/IBM_data.gguf

    -l models/kore_lora/lora-IBM-LATEST.bin

for finishing finetuning. My llama.cpp-master only creates .gguf checkpoint files after finetuning, there is no .bin file…

Is this file so important or can I do export-lora without it?
Is there some way to convert .gguf file into .bin separately?
Or maybe someone has the same problem?