English
sft
StableLM
Edit model card

Exllama v2 Quantizations of Nous-Capybara-7B-V1.9

Using turboderp's ExLlamaV2 v0.0.7 for quantization.

Each branches contains an individual bits per weight.

Conversion was done using wikitext.parquet as calibration dataset.

Original model: https://huggingface.co/NousResearch/Nous-Capybara-7B-V1.9

4.0 bits per weight

6.0 bits per weight

8.0 bits per weight

Download instructions

With git:

git clone --single-branch --branch 4.0 https://huggingface.co/bartowski/Nous-Capybara-7B-V1.9-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 Nous-Capybara-7B-V1.9-exl2:

mkdir Nous-Capybara-7B-V1.9-exl2
huggingface-cli download bartowski/Nous-Capybara-7B-V1.9-exl2 --local-dir Nous-Capybara-7B-V1.9-exl2 --local-dir-use-symlinks False

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

mkdir Nous-Capybara-7B-V1.9-exl2
huggingface-cli download bartowski/Nous-Capybara-7B-V1.9-exl2 --revision 4.0 --local-dir Nous-Capybara-7B-V1.9-exl2 --local-dir-use-symlinks False
Downloads last month
0
Unable to determine this model's library. Check the docs .

Datasets used to train bartowski/Nous-Capybara-7B-V1.9-exl2