Orca-2-13b-exl2 / README.md
bartowski's picture
Quant for 4.0
f449057
|
raw
history blame
906 Bytes
metadata
pipeline_tag: text-generation
tags:
  - orca
  - orca2
  - microsoft
quantized_by: bartowski

Exllama v2 Quantizations of Orca-2-13b at 4.0 bits per weight

Using turboderp's ExLlamaV2 v0.0.8 for quantization.

Conversion was done using wikitext-103-raw-v1-test.parquet as calibration dataset.

Original model: https://huggingface.co/microsoft/Orca-2-13b

Download instructions

With git:

git clone --single-branch --branch 4.0 https://huggingface.co/bartowski/Orca-2-13b-exl2

With huggingface hub (credit to TheBloke for instructions):

pip3 install huggingface-hub

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

mkdir Orca-2-13b-exl2
huggingface-cli download bartowski/Orca-2-13b-exl2 --revision 4_0 --local-dir Orca-2-13b-exl2 --local-dir-use-symlinks False