I tried using the mergekit.
I used this provided yaml file as a template, just changed the model names.
models:
- model: mistralai/Mistral-7B-v0.1
#no parameters necessary for base model - model: mistralai/Mistral-7B-Instruct-v0.2
parameters:
density: 0.5
weight: 0.5 - model: BioMistral/BioMistral-7B
parameters:
density: 0.5
weight: 0.5
merge_method: ties
base_model: mistralai/Mistral-7B-v0.1
parameters:
normalize: false
int8_mask: true
dtype: float16
It did not work and just said “error”.
I did leave this field empty:
HF Write Token
How do tokens work, do you need one to use the mergekit ?
Was the error because of the token or something else ?
Can you make a model straight to GGUF instead of float 16 ?