English
bartowski's picture
Quant for 6.0
8d15860
metadata
datasets:
  - ehartford/dolphin
  - jondurbin/airoboros-2.2.1
  - ehartford/samantha-data
  - ehartford/WizardLM_evol_instruct_V2_196k_unfiltered_merged_split
language:
  - en
license: other
license_name: yi-license
license_link: LICENSE
quantized_by: bartowski

Exllama v2 Quantizations of dolphin-2_2-yi-34b at 6.0,4.25,4.0,3.75

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/ehartford/dolphin-2_2-yi-34b

Download instructions

With git:

git clone --single-branch --branch 6.0,4.25,4.0,3.75 https://huggingface.co/bartowski/dolphin-2_2-yi-34b-exl2

With huggingface hub (credit to TheBloke for instructions):

pip3 install huggingface-hub

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

mkdir dolphin-2_2-yi-34b-exl2
huggingface-cli download bartowski/dolphin-2_2-yi-34b-exl2 --revision  --local-dir dolphin-2_2-yi-34b-exl2 --local-dir-use-symlinks False