Exllama v2 Quantizations of airoboros-2.2.1-y34b
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/Doctor-Shotgun/airoboros-2.2.1-y34b
Download instructions
With git:
git clone --single-branch --branch 4.0 https://huggingface.co/bartowski/airoboros-2.2.1-y34b-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 airoboros-2.2.1-y34b-exl2:
mkdir airoboros-2.2.1-y34b-exl2
huggingface-cli download bartowski/airoboros-2.2.1-y34b-exl2 --local-dir airoboros-2.2.1-y34b-exl2 --local-dir-use-symlinks False
To download from a different branch, add the --revision parameter:
mkdir airoboros-2.2.1-y34b-exl2
huggingface-cli download bartowski/airoboros-2.2.1-y34b-exl2 --revision 4.0 --local-dir airoboros-2.2.1-y34b-exl2 --local-dir-use-symlinks False