CatPPT-exl2 / README.md
bartowski's picture
Add 4.0 link
a097dd2
|
raw
history blame
2.46 kB
---
license: apache-2.0
library_name: peft
base_model: rishiraj/CatPPT-base
datasets:
- HuggingFaceH4/no_robots
language:
- en
widget:
- text: |
<|system|>
You are a friendly chatbot who always responds in the style of a pirate</s>
<|user|>
How many helicopters can a human eat in one sitting?</s>
<|assistant|>
output:
text: >-
Aye, me hearties! 'Tis not likely a human can eat a helicopter in any sittin', let alone one! They be too big and made of metal, and not fit for consumption. But if ye be referrin' to helicopter snacks, like nuts and trail mix, then a human might be able to munch a goodly amount in one sittin'. Arr!
tags:
- generated_from_trainer
pipeline_tag: text-generation
model-index:
- name: CatPPT
results: []
quantized_by: bartowski
---
## Exllama v2 Quantizations of CatPPT
Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.11">turboderp's ExLlamaV2 v0.0.11</a> for quantization.
Each branch contains an individual bits per weight, with the main one containing only the meaurement.json for further conversions.
Conversion was done using the default calibration dataset.
Default arguments used except when the bits per weight is above 6.0, at that point the lm_head layer is quantized at 8 bits per weight instead of the default 6.
Original model: https://huggingface.co/rishiraj/CatPPT
<a href="https://huggingface.co/bartowski/CatPPT-exl2/tree/4_0">4.0 bits per weight</a>
<a href="https://huggingface.co/bartowski/CatPPT-exl2/tree/5_0">5.0 bits per weight</a>
<a href="https://huggingface.co/bartowski/CatPPT-exl2/tree/6_5">6.5 bits per weight</a>
<a href="https://huggingface.co/bartowski/CatPPT-exl2/tree/8_0">8.0 bits per weight</a>
## Download instructions
With git:
```shell
git clone --single-branch --branch 4_0 https://huggingface.co/bartowski/CatPPT-exl2
```
With huggingface hub (credit to TheBloke for instructions):
```shell
pip3 install huggingface-hub
```
To download the `main` (only useful if you only care about measurement.json) branch to a folder called `CatPPT-exl2`:
```shell
mkdir CatPPT-exl2
huggingface-cli download bartowski/CatPPT-exl2 --local-dir CatPPT-exl2 --local-dir-use-symlinks False
```
To download from a different branch, add the `--revision` parameter:
```shell
mkdir CatPPT-exl2
huggingface-cli download bartowski/CatPPT-exl2 --revision 4_0 --local-dir CatPPT-exl2 --local-dir-use-symlinks False
```