README.md

metadata

quantized_by: bartowski

Exllama v2 Quantizations of bagel-dpo-7b-v0.1 at 4.0 bits per weight

Using turboderp's ExLlamaV2 v0.0.10 for quantization.

Conversion was done using VMWareOpenInstruct.parquet as calibration dataset.

Original model: https://huggingface.co/jondurbin/bagel-dpo-7b-v0.1

Download instructions

With git:

git clone --single-branch --branch 4.0 https://huggingface.co/bartowski/bagel-dpo-7b-v0.1-exl2

With huggingface hub (credit to TheBloke for instructions):

pip3 install huggingface-hub

To download from a different branch, add the --revision parameter:

mkdir bagel-dpo-7b-v0.1-exl2
huggingface-cli download bartowski/bagel-dpo-7b-v0.1-exl2 --revision 4_0 --local-dir bagel-dpo-7b-v0.1-exl2 --local-dir-use-symlinks False