Updating the GGUF files
#4
by
Armada22
- opened
Are you planning on using the newest version of llama.cpp that implements the fixes from https://github.com/ggerganov/llama.cpp/pull/6920 to update the files?
Yes, but later!
No worry!
Alright, I tried to do it, here's my issue :
- If I use convert.py => don't work
- If I use convert-hf-to-gguf.py => don't work
- If I use convert.py with --vocab-type bpe => work but the outputed files are the same
Hum...
I didn't used the tokenizer.model from Llama3 (only the tokenizer.json) and I didn't got any issue with the token myself even when people told us they have... So I dunno, please report if you have issue.
I currently fix all the model, this one is the next in some minutes. Have fun!
Thank you so much!
Armada22
changed discussion status to
closed