README.md

---
license: other
language:
- en
pipeline_tag: text2text-generation
tags:
- alpaca
- llama
- chat
- gpt4
inference: false
---
# GPT4 Alpaca LoRA 30B - 4bit GGML

This is a 4-bit GGML version of the [Chansung GPT4 Alpaca 30B LoRA model](https://huggingface.co/chansung/gpt4-alpaca-lora-30b).

It was created by merging the LoRA provided in the above repo with the original Llama 30B model, producing unquantised model [GPT4-Alpaca-LoRA-30B-HF](https://huggingface.co/TheBloke/gpt4-alpaca-lora-30b-HF)

The files in this repo were then quantized to 4bit and 5bit for use with [llama.cpp](https://github.com/ggerganov/llama.cpp).

## Provided files
| Name | Quant method | Bits | Size | RAM required | Use case |
| ---- | ---- | ---- | ---- | ---- | ----- |
`gpt4-alpaca-lora-30B.GGML.q4_0.bin` | q4_0 | 4bit | 20.3GB | 23GB | Maximum compatibility |
`gpt4-alpaca-lora-30B.GGML.q4_2.bin` | q4_2 | 4bit | 20.3GB | 23GB | Best compromise between resources, speed and quality |
`gpt4-alpaca-lora-30B.GGML.q5_0.bin` | q5_0 | 5bit | 22.4GB | 25GB | Brand new 5bit method. Potentially higher quality than 4bit, at cost of slightly higher resources. |
`gpt4-alpaca-lora-30B.GGML.q5_1.bin` | q5_1 | 5bit | 24.4GB | 27GB | Brand new 5bit method. Slightly higher resource usage than q5_0.|

* The q4_0 file provides lower quality, but maximal compatibility. It will work with past and future versions of llama.cpp* The q4_0 file provides lower quality, but maximal compatibility. It will work with past and future versions of llama.cpp
* The q4_2 file offers the best combination of performance and quality. This format is still subject to change and there may be compatibility issues, see below.
* The q5_0 file is using brand new 5bit method released 26th April. This is the 5bit equivalent of q4_0.
* The q5_1 file is using brand new 5bit method released 26th April. This is the 5bit equivalent of q4_1.

## q4_2 compatibility

q4_2 is a relatively new 4bit quantisation method offering improved quality. However they are still under development and their formats are subject to change.

In order to use these files you will need to use recent llama.cpp code. And it's possible that future updates to llama.cpp could require that these files are re-generated.

If and when the q4_2 file no longer works with recent versions of llama.cpp I will endeavour to update it.

If you want to ensure guaranteed compatibility with a wide range of llama.cpp versions, use the q4_0 file.

## q5_0 and q5_1 compatibility

These new methods were released to llama.cpp on 26th April. You will need to pull the latest llama.cpp code and rebuild to be able to use them.

Third-party UIs/tools may not support this yet.

## How to run in `llama.cpp`

I use the following command line; adjust for your tastes and needs:

```
./main -t 18 -m gpt4-alpaca-lora-30B.GGML.q4_2.bin --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
Write a story about llamas
### Response:"
```
Change `-t 18` to the number of physical CPU cores you have. For example if your system has 8 cores/16 threads, use `-t 8`.

If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`

## How to run in `text-generation-webui`

Create a model directory that has  `ggml` (case sensitive) in its name. Then put the desired .bin file in that model directory.

Further instructions here: [text-generation-webui/docs/llama.cpp-models.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/llama.cpp-models.md).

# Original GPT4 Alpaca Lora model card

This repository comes with LoRA checkpoint to make LLaMA into a chatbot like language model. The checkpoint is the output of instruction following fine-tuning process with the following settings on 8xA100(40G) DGX system.
- Training script: borrowed from the official [Alpaca-LoRA](https://github.com/tloen/alpaca-lora) implementation
- Training script:
```shell
python finetune.py \
    --base_model='decapoda-research/llama-30b-hf' \
    --data_path='alpaca_data_gpt4.json' \
    --num_epochs=10 \
    --cutoff_len=512 \
    --group_by_length \
    --output_dir='./gpt4-alpaca-lora-30b' \
    --lora_target_modules='[q_proj,k_proj,v_proj,o_proj]' \
    --lora_r=16 \
    --batch_size=... \
    --micro_batch_size=...
```

You can find how the training went from W&B report [here](https://wandb.ai/chansung18/gpt4_alpaca_lora/runs/w3syd157?workspace=user-chansung18).