bartowski/FusionNet_7Bx2_MoE_14B-exl2

Exllama v2 Quantizations of FusionNet_7Bx2_MoE_14B

Using turboderp's ExLlamaV2 v0.0.11 for quantization.

The "main" branch only contains the measurement.json, download one of the other branches for the model (see below)

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/TomGrc/FusionNet_7Bx2_MoE_14B

8.0 bits per weight

6.5 bits per weight

5.0 bits per weight

4.0 bits per weight

3.5 bits per weight

Download instructions

With git:

git clone --single-branch --branch 4_0 https://huggingface.co/bartowski/FusionNet_7Bx2_MoE_14B-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 FusionNet_7Bx2_MoE_14B-exl2:

mkdir FusionNet_7Bx2_MoE_14B-exl2
huggingface-cli download bartowski/FusionNet_7Bx2_MoE_14B-exl2 --local-dir FusionNet_7Bx2_MoE_14B-exl2 --local-dir-use-symlinks False

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

mkdir FusionNet_7Bx2_MoE_14B-exl2
huggingface-cli download bartowski/FusionNet_7Bx2_MoE_14B-exl2 --revision 4_0 --local-dir FusionNet_7Bx2_MoE_14B-exl2 --local-dir-use-symlinks False