Exllama v2 Quantizations of Orca-2-7b
Using turboderp's ExLlamaV2 v0.0.8 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.
Default arguments used except when the bits per weight is above 6.0, at that point the lm_head layer is quantized at 8 bits per weight instead of the default 6.
Original model: https://huggingface.co/microsoft/Orca-2-7b
Download instructions
With git:
git clone --single-branch --branch 4_0 https://huggingface.co/bartowski/Orca-2-7b-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 Orca-2-7b-exl2:
mkdir Orca-2-7b-exl2
huggingface-cli download bartowski/Orca-2-7b-exl2 --local-dir Orca-2-7b-exl2 --local-dir-use-symlinks False
To download from a different branch, add the --revision parameter:
mkdir Orca-2-7b-exl2
huggingface-cli download bartowski/Orca-2-7b-exl2 --revision 4_0 --local-dir Orca-2-7b-exl2 --local-dir-use-symlinks False