Synthia-v3.0-11B-exl2

Exllama v2 Quantizations of Synthia-v3.0-11B

Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.11">turboderp's ExLlamaV2 v0.0.11</a> 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 the default 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/migtissera/Synthia-v3.0-11B

<a href="https://huggingface.co/bartowski/Synthia-v3.0-11B-exl2/tree/3_75">3.75 bits per weight</a>

<a href="https://huggingface.co/bartowski/Synthia-v3.0-11B-exl2/tree/4_0">4.0 bits per weight</a>

<a href="https://huggingface.co/bartowski/Synthia-v3.0-11B-exl2/tree/5_0">5.0 bits per weight</a>

<a href="https://huggingface.co/bartowski/Synthia-v3.0-11B-exl2/tree/6_5">6.5 bits per weight</a>

Download instructions

With git:

git clone --single-branch --branch 4_0 https://huggingface.co/bartowski/Synthia-v3.0-11B-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 Synthia-v3.0-11B-exl2:

mkdir Synthia-v3.0-11B-exl2
huggingface-cli download bartowski/Synthia-v3.0-11B-exl2 --local-dir Synthia-v3.0-11B-exl2 --local-dir-use-symlinks False

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

mkdir Synthia-v3.0-11B-exl2
huggingface-cli download bartowski/Synthia-v3.0-11B-exl2 --revision 4_0 --local-dir Synthia-v3.0-11B-exl2 --local-dir-use-symlinks False