mradermacher

/

magnum-72b-v1-GGUF

like 0

I downloaded the Q4_K_S version, and llama.cpp shows this:
...
llm_load_print_meta: general.name = magnum-72b-v1
llm_load_print_meta: BOS token = 151645 '<|im_end|>'
llm_load_print_meta: EOS token = 151645 '<|im_end|>'
llm_load_print_meta: PAD token = 151643 '<|endoftext|>'
llm_load_print_meta: LF token = 148848 'ÄĬ'
llm_load_print_meta: EOT token = 151645 '<|im_end|>'
...
Is the BOS token correct, or it's just this model have unique settings? Thanks.