|
--- |
|
library_name: transformers |
|
language: |
|
- en |
|
- fr |
|
- de |
|
- es |
|
- it |
|
- pt |
|
- ja |
|
- ko |
|
- zh |
|
- ar |
|
- el |
|
- fa |
|
- pl |
|
- id |
|
- cs |
|
- he |
|
- hi |
|
- nl |
|
- ro |
|
- ru |
|
- tr |
|
- uk |
|
- vi |
|
license: cc-by-nc-4.0 |
|
quantized_by: bartowski |
|
pipeline_tag: text-generation |
|
--- |
|
|
|
## Exllama v2 Quantizations of aya-23-8B |
|
|
|
Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.21">turboderp's ExLlamaV2 v0.0.21</a> for quantization. |
|
|
|
<b>The "main" branch only contains the measurement.json, download one of the other branches for the model (see below)</b> |
|
|
|
Each branch contains an individual bits per weight, with the main one containing only the meaurement.json for further conversions. |
|
|
|
Original model: https://huggingface.co/CohereForAI/aya-23-8B |
|
|
|
## Prompt format |
|
|
|
``` |
|
<BOS_TOKEN><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>{system_prompt}<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>{prompt}<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|><|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|> |
|
``` |
|
|
|
## Available sizes |
|
|
|
|
|
| Branch | Bits | lm_head bits | VRAM (4k) | VRAM (8K) | VRAM (16k) | VRAM (32k) | Description | |
|
| ----- | ---- | ------- | ------ | ------ | ------ | ------ | ------------ | |
|
| [8_0](https://huggingface.co/bartowski/aya-23-8B-exl2/tree/8_0) | 8.0 | 8.0 | 10.1 GB | 10.5 GB | 11.5 GB | 13.6 GB | Maximum quality that ExLlamaV2 can produce, near unquantized performance. | |
|
| [6_5](https://huggingface.co/bartowski/aya-23-8B-exl2/tree/6_5) | 6.5 | 8.0 | 8.9 GB | 9.3 GB | 10.3 GB | 12.4 GB | Very similar to 8.0, good tradeoff of size vs performance, **recommended**. | |
|
| [5_0](https://huggingface.co/bartowski/aya-23-8B-exl2/tree/5_0) | 5.0 | 6.0 | 7.7 GB | 8.1 GB | 9.1 GB | 11.2 GB | Slightly lower quality vs 6.5, but usable on 8GB cards. | |
|
| [4_25](https://huggingface.co/bartowski/aya-23-8B-exl2/tree/4_25) | 4.25 | 6.0 | 7.0 GB | 7.4 GB | 8.4 GB | 10.5 GB | GPTQ equivalent bits per weight, slightly higher quality. | |
|
| [3_5](https://huggingface.co/bartowski/aya-23-8B-exl2/tree/3_5) | 3.5 | 6.0 | 6.4 GB | 6.8 GB | 7.8 GB | 9.9 GB | Lower quality, only use if you have to. | |
|
|
|
## Download instructions |
|
|
|
With git: |
|
|
|
```shell |
|
git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/aya-23-8B-exl2 aya-23-8B-exl2-6_5 |
|
``` |
|
|
|
With huggingface hub (credit to TheBloke for instructions): |
|
|
|
```shell |
|
pip3 install huggingface-hub |
|
``` |
|
|
|
To download a specific branch, use the `--revision` parameter. For example, to download the 6.5 bpw branch: |
|
|
|
Linux: |
|
|
|
```shell |
|
huggingface-cli download bartowski/aya-23-8B-exl2 --revision 6_5 --local-dir aya-23-8B-exl2-6_5 |
|
``` |
|
|
|
Windows (which apparently doesn't like _ in folders sometimes?): |
|
|
|
```shell |
|
huggingface-cli download bartowski/aya-23-8B-exl2 --revision 6_5 --local-dir aya-23-8B-exl2-6.5 |
|
``` |
|
|
|
Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski |
|
|
