bartowski's picture
Quant for 4.0
8cbbdbf
metadata
tags:
  - llm
  - llama
  - llama2
quantized_by: bartowski

Exllama v2 Quantizations of Free_Sydney_V2_Mistral_7b 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/FPHam/Free_Sydney_V2_Mistral_7b

Download instructions

With git:

git clone --single-branch --branch 4.0 https://huggingface.co/bartowski/Free_Sydney_V2_Mistral_7b-exl2

With huggingface hub (credit to TheBloke for instructions):

pip3 install huggingface-hub

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

mkdir Free_Sydney_V2_Mistral_7b-exl2
huggingface-cli download bartowski/Free_Sydney_V2_Mistral_7b-exl2 --revision 4_0 --local-dir Free_Sydney_V2_Mistral_7b-exl2 --local-dir-use-symlinks False