Exllama v2 Quantizations of Yarn-Mistral-7b-128k
Using turboderp's ExLlamaV2 v0.0.7 for quantization.
Each branch contains an individual bits per weight, with the main one containing only the meaurement.json for further conversions.
Conversion was done using wikitext-103-raw-v1-test.parquet as calibration dataset.
Original model: https://huggingface.co/NousResearch/Yarn-Mistral-7b-128k
Download instructions
With git:
git clone --single-branch --branch 4.0 https://huggingface.co/bartowski/Yarn-Mistral-7b-128k-exl2
With huggingface hub (credit to TheBloke for instructions):
pip3 install huggingface-hub
To download the main (only useful if you only care about measurement.json) branch to a folder called Yarn-Mistral-7b-128k-exl2:
mkdir Yarn-Mistral-7b-128k-exl2
huggingface-cli download bartowski/Yarn-Mistral-7b-128k-exl2 --local-dir Yarn-Mistral-7b-128k-exl2 --local-dir-use-symlinks False
To download from a different branch, add the --revision parameter:
mkdir Yarn-Mistral-7b-128k-exl2
huggingface-cli download bartowski/Yarn-Mistral-7b-128k-exl2 --revision 4.0 --local-dir Yarn-Mistral-7b-128k-exl2 --local-dir-use-symlinks False