This is the test version for pruning. This model is a base model that will be pruned and quantized for on-device purpose.
I used mergekit for merging two models:
https://github.com/cg123/mergekit The two models I combined are:
https://huggingface.co/jeonsworld/CarbonVillain-en-10.7B-v2 https://huggingface.co/kyujinpy/Sakura-SOLAR-Instruct-DPO-v2
I used GGUF quantization.