Is this compatible with GGML?
#4
by
Laurab
- opened
I'm trying to convert this model to GGML but I'm getting
$ python convert.py models/llama-2-orca
...
Exception: Vocab size mismatch (model has 32016, but models/tokenizer.model combined with models/added_tokens.json has 32006).
Maybe this is of interest..
https://huggingface.co/OpenAssistant/oasst-sft-6-llama-30b-xor/discussions/2