Text Generation
Transformers
English
Inference Endpoints
bartowski's picture
Quant for 3.75
84bbb15
|
raw
history blame
1.06 kB
---
license: apache-2.0
datasets:
- teknium/openhermes
- argilla/ultrafeedback-binarized-preferences
- Intel/orca_dpo_pairs
language:
- en
library_name: transformers
quantized_by: bartowski
---
# Exllama v2 Quantizations of DPOpenHermes-11B at 3.75 bits per weight
Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.10">turboderp's ExLlamaV2 v0.0.10</a> for quantization.
Conversion was done using wikitext-103-raw-v1-test.parquet as calibration dataset.
Original model: https://huggingface.co/openaccess-ai-collective/DPOpenHermes-11B
## Download instructions
With git:
```shell
git clone --single-branch --branch 3.75 https://huggingface.co/bartowski/DPOpenHermes-11B-exl2
```
With huggingface hub (credit to TheBloke for instructions):
```shell
pip3 install huggingface-hub
```
To download from a different branch, add the `--revision` parameter:
```shell
mkdir DPOpenHermes-11B-exl2
huggingface-cli download bartowski/DPOpenHermes-11B-exl2 --revision 3_75 --local-dir DPOpenHermes-11B-exl2 --local-dir-use-symlinks False
```