---
language:
- eng
tags:
- sft
- StableLM
license:
- mit
datasets:
- LDJnr/LessWrong-Amplify-Instruct
- LDJnr/Pure-Dove
- LDJnr/Verified-Camel
quantized_by: bartowski
---

# Exllama v2 Quantizations of Nous-Capybara-34B at 8.0,6.0,4.25,4.0,3.75

Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.8">turboderp's ExLlamaV2 v0.0.8</a> for quantization.

Conversion was done using wikitext-103-raw-v1-test.parquet as calibration dataset.

Original model: https://huggingface.co./NousResearch/Nous-Capybara-34B

## Download instructions

With git:

```shell
git clone --single-branch --branch 8.0,6.0,4.25,4.0,3.75 https://huggingface.co./bartowski/Nous-Capybara-34B-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 Nous-Capybara-34B-exl2
huggingface-cli download bartowski/Nous-Capybara-34B-exl2 --revision  --local-dir Nous-Capybara-34B-exl2 --local-dir-use-symlinks False
```