TheBloke/MonadGPT-GPTQ

Chat & support: TheBloke's Discord server

Want to contribute? TheBloke's Patreon page

TheBloke's LLM work is generously supported by a grant from andreessen horowitz (a16z)

---

MonadGPT 7B - GPTQ

• Model creator: Pierre-Carl Langlais
• Original model: MonadGPT 7B

Description

This repo contains GPTQ model files for Pierre-Carl Langlais's MonadGPT 7B.

Multiple GPTQ parameter permutations are provided; see Provided Files below for details of the options provided, their parameters, and the software used to create them.

These files were quantised using hardware kindly provided by Massed Compute.

Repositories available

• AWQ model(s) for GPU inference.
• GPTQ models for GPU inference, with multiple quantisation parameter options.
• 2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference
• Pierre-Carl Langlais's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions

Prompt template: ChatML

<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

Known compatible clients / servers

These GPTQ models are known to work in the following inference servers/webuis.

• text-generation-webui
• KoboldAI United
• LoLLMS Web UI
• Hugging Face Text Generation Inference (TGI)

This may not be a complete list; if you know of others, please let me know!

Provided files, and GPTQ parameters

Multiple quantisation parameters are provided, to allow you to choose the best one for your hardware and requirements.

Each separate quant is in a different branch. See below for instructions on fetching from different branches.

Most GPTQ files are made with AutoGPTQ. Mistral models are currently made with Transformers.

Explanation of GPTQ parameters

• Bits: The bit size of the quantised model.
• GS: GPTQ group size. Higher numbers use less VRAM, but have lower quantisation accuracy. "None" is the lowest possible value.
• Act Order: True or False. Also known as desc_act. True results in better quantisation accuracy. Some GPTQ clients have had issues with models that use Act Order plus Group Size, but this is generally resolved now.
• Damp %: A GPTQ parameter that affects how samples are processed for quantisation. 0.01 is default, but 0.1 results in slightly better accuracy.
• GPTQ dataset: The calibration dataset used during quantisation. Using a dataset more appropriate to the model's training can improve quantisation accuracy. Note that the GPTQ calibration dataset is not the same as the dataset used to train the model - please refer to the original model repo for details of the training dataset(s).
• Sequence Length: The length of the dataset sequences used for quantisation. Ideally this is the same as the model sequence length. For some very long sequence models (16+K), a lower sequence length may have to be used. Note that a lower sequence length does not limit the sequence length of the quantised model. It only impacts the quantisation accuracy on longer inference sequences.
• ExLlama Compatibility: Whether this file can be loaded with ExLlama, which currently only supports Llama and Mistral models in 4-bit.

Branch	Bits	GS	Act Order	Damp %	GPTQ Dataset	Seq Len	Size	ExLlama	Desc
main	4	128	Yes	0.1	latin-english	4096	4.16 GB	Yes	4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy.
gptq-4bit-32g-actorder_True	4	32	Yes	0.1	latin-english	4096	4.57 GB	Yes	4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage.
gptq-8bit--1g-actorder_True	8	None	Yes	0.1	latin-english	4096	7.52 GB	No	8-bit, with Act Order. No group size, to lower VRAM requirements.
gptq-8bit-128g-actorder_True	8	128	Yes	0.1	latin-english	4096	7.68 GB	No	8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy.
gptq-8bit-32g-actorder_True	8	32	Yes	0.1	latin-english	4096	8.17 GB	No	8-bit, with group size 32g and Act Order for maximum inference quality.
gptq-4bit-64g-actorder_True	4	64	Yes	0.1	latin-english	4096	4.30 GB	Yes	4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy.

How to download, including from branches

In text-generation-webui

To download from the main branch, enter TheBloke/MonadGPT-GPTQ in the "Download model" box.

To download from another branch, add :branchname to the end of the download name, eg TheBloke/MonadGPT-GPTQ:gptq-4bit-32g-actorder_True

From the command line

I recommend using the huggingface-hub Python library:

pip3 install huggingface-hub

To download the main branch to a folder called MonadGPT-GPTQ:

mkdir MonadGPT-GPTQ
huggingface-cli download TheBloke/MonadGPT-GPTQ --local-dir MonadGPT-GPTQ --local-dir-use-symlinks False

To download from a different branch, add the --revision parameter:

mkdir MonadGPT-GPTQ
huggingface-cli download TheBloke/MonadGPT-GPTQ --revision gptq-4bit-32g-actorder_True --local-dir MonadGPT-GPTQ --local-dir-use-symlinks False

More advanced huggingface-cli download usage

If you remove the --local-dir-use-symlinks False parameter, the files will instead be stored in the central Hugging Face cache directory (default location on Linux is: ~/.cache/huggingface), and symlinks will be added to the specified --local-dir, pointing to their real location in the cache. This allows for interrupted downloads to be resumed, and allows you to quickly clone the repo to multiple places on disk without triggering a download again. The downside, and the reason why I don't list that as the default option, is that the files are then hidden away in a cache folder and it's harder to know where your disk space is being used, and to clear it up if/when you want to remove a download model.

The cache location can be changed with the HF_HOME environment variable, and/or the --cache-dir parameter to huggingface-cli.

For more documentation on downloading with huggingface-cli, please see: HF -> Hub Python Library -> Download files -> Download from the CLI.

To accelerate downloads on fast connections (1Gbit/s or higher), install hf_transfer:

pip3 install hf_transfer

And set environment variable HF_HUB_ENABLE_HF_TRANSFER to 1:

mkdir MonadGPT-GPTQ
HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/MonadGPT-GPTQ --local-dir MonadGPT-GPTQ --local-dir-use-symlinks False

Windows Command Line users: You can set the environment variable by running set HF_HUB_ENABLE_HF_TRANSFER=1 before the download command.

With git (not recommended)

To clone a specific branch with git, use a command like this:

git clone --single-branch --branch gptq-4bit-32g-actorder_True https://huggingface.co/TheBloke/MonadGPT-GPTQ

Note that using Git with HF repos is strongly discouraged. It will be much slower than using huggingface-hub, and will use twice as much disk space as it has to store the model files twice (it stores every byte both in the intended target folder, and again in the .git folder as a blob.)

How to easily download and use this model in text-generation-webui

Please make sure you're using the latest version of text-generation-webui.

It is strongly recommended to use the text-generation-webui one-click-installers unless you're sure you know how to make a manual install.

1. Click the Model tab.

2. Under Download custom model or LoRA, enter TheBloke/MonadGPT-GPTQ.

   • To download from a specific branch, enter for example TheBloke/MonadGPT-GPTQ:gptq-4bit-32g-actorder_True
   • see Provided Files above for the list of branches for each option.

3. Click Download.

4. The model will start downloading. Once it's finished it will say "Done".

5. In the top left, click the refresh icon next to Model.

6. In the Model dropdown, choose the model you just downloaded: MonadGPT-GPTQ

7. The model will automatically load, and is now ready for use!

8. If you want any custom settings, set them and then click Save settings for this model followed by Reload the Model in the top right.

   • Note that you do not need to and should not set manual GPTQ parameters any more. These are set automatically from the file quantize_config.json.

9. Once you're ready, click the Text Generation tab and enter a prompt to get started!

Serving this model from Text Generation Inference (TGI)

It's recommended to use TGI version 1.1.0 or later. The official Docker container is: ghcr.io/huggingface/text-generation-inference:1.1.0

Example Docker parameters:

--model-id TheBloke/MonadGPT-GPTQ --port 3000 --quantize gptq --max-input-length 3696 --max-total-tokens 4096 --max-batch-prefill-tokens 4096

Example Python code for interfacing with TGI (requires huggingface-hub 0.17.0 or later):

pip3 install huggingface-hub

from huggingface_hub import InferenceClient

endpoint_url = "https://your-endpoint-url-here"

prompt = "Tell me about AI"
prompt_template=f'''<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
'''

client = InferenceClient(endpoint_url)
response = client.text_generation(prompt,
                                  max_new_tokens=128,
                                  do_sample=True,
                                  temperature=0.7,
                                  top_p=0.95,
                                  top_k=40,
                                  repetition_penalty=1.1)

print(f"Model output: {response}")

How to use this GPTQ model from Python code

Install the necessary packages

Requires: Transformers 4.33.0 or later, Optimum 1.12.0 or later, and AutoGPTQ 0.4.2 or later.

pip3 install transformers optimum
pip3 install auto-gptq --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/  # Use cu117 if on CUDA 11.7

If you have problems installing AutoGPTQ using the pre-built wheels, install it from source instead:

pip3 uninstall -y auto-gptq
git clone https://github.com/PanQiWei/AutoGPTQ
cd AutoGPTQ
git checkout v0.4.2
pip3 install .

You can then use the following code

from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline

model_name_or_path = "TheBloke/MonadGPT-GPTQ"
# To use a different branch, change revision
# For example: revision="gptq-4bit-32g-actorder_True"
model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
                                             device_map="auto",
                                             trust_remote_code=False,
                                             revision="main")

tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)

prompt = "Tell me about AI"
prompt_template=f'''<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
'''

print("\n\n*** Generate:")

input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
output = model.generate(inputs=input_ids, temperature=0.7, do_sample=True, top_p=0.95, top_k=40, max_new_tokens=512)
print(tokenizer.decode(output[0]))

# Inference can also be done using transformers' pipeline

print("*** Pipeline:")
pipe = pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    max_new_tokens=512,
    do_sample=True,
    temperature=0.7,
    top_p=0.95,
    top_k=40,
    repetition_penalty=1.1
)

print(pipe(prompt_template)[0]['generated_text'])

Compatibility

The files provided are tested to work with Transformers. For non-Mistral models, AutoGPTQ can also be used directly.

ExLlama is compatible with Llama and Mistral models in 4-bit. Please see the Provided Files table above for per-file compatibility.

For a list of clients/servers, please see "Known compatible clients / servers", above.

Discord

For further support, and discussions on these models and AI in general, join us at:

TheBloke AI's Discord server

Thanks, and how to contribute

Thanks to the chirper.ai team!

Thanks to Clay from gpus.llm-utils.org!

I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.

If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.

Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.

• Patreon: https://patreon.com/TheBlokeAI
• Ko-Fi: https://ko-fi.com/TheBlokeAI

Special thanks to: Aemon Algiz.

Patreon special mentions: Brandon Frisco, LangChain4j, Spiking Neurons AB, transmissions 11, Joseph William Delisle, Nitin Borwankar, Willem Michiel, Michael Dempsey, vamX, Jeffrey Morgan, zynix, jjj, Omer Bin Jawed, Sean Connelly, jinyuan sun, Jeromy Smith, Shadi, Pawan Osman, Chadd, Elijah Stavena, Illia Dulskyi, Sebastain Graf, Stephen Murray, terasurfer, Edmond Seymore, Celu Ramasamy, Mandus, Alex, biorpg, Ajan Kanaga, Clay Pascal, Raven Klaugh, 阿明, K, ya boyyy, usrbinkat, Alicia Loh, John Villwock, ReadyPlayerEmma, Chris Smitley, Cap'n Zoog, fincy, GodLy, S_X, sidney chen, Cory Kujawski, OG, Mano Prime, AzureBlack, Pieter, Kalila, Spencer Kim, Tom X Nguyen, Stanislav Ovsiannikov, Michael Levine, Andrey, Trailburnt, Vadim, Enrico Ros, Talal Aujan, Brandon Phillips, Jack West, Eugene Pentland, Michael Davis, Will Dee, webtim, Jonathan Leane, Alps Aficionado, Rooh Singh, Tiffany J. Kim, theTransient, Luke @flexchar, Elle, Caitlyn Gatomon, Ari Malik, subjectnull, Johann-Peter Hartmann, Trenton Dambrowitz, Imad Khwaja, Asp the Wyvern, Emad Mostaque, Rainer Wilmers, Alexandros Triantafyllidis, Nicholas, Pedro Madruga, SuperWojo, Harry Royden McLaughlin, James Bentley, Olakabola, David Ziegler, Ai Maven, Jeff Scroggin, Nikolai Manek, Deo Leter, Matthew Berman, Fen Risland, Ken Nordquist, Manuel Alberto Morcote, Luke Pendergrass, TL, Fred von Graf, Randy H, Dan Guido, NimbleBox.ai, Vitor Caleffi, Gabriel Tamborski, knownsqashed, Lone Striker, Erik Bjäreholt, John Detwiler, Leonard Tan, Iucharbius

Thank you to all my generous patrons and donaters!

And thank you again to a16z for their generous grant.

Original model card: Pierre-Carl Langlais's MonadGPT 7B

What would have happened if ChatGPT was invented in the 17th century? MonadGPT is a possible answer.

MonadGPT is a finetune of Mistral-Hermes on 11,000 early modern texts in English, French and Latin, mostly coming from EEBO and Gallica.

Like the original Mistral-Hermes, MonadGPT can be used in conversation mode. It will not only answer in an historical language and style but will use historical and dated references. This is especially visible for science questions (astronomy, medecine). Obviously, it's not recommended to follow any advice from Monad-GPT

Examples

User: What are the planets of the solar system?
Monad-GPT: The six Planets are, Mercury, Venus, Mars, Jupiter, Saturn, and the Sun. The Moon is also a Planet; which being of so great Influence upon Earth, makes it to be most admired of all others by Astronomers.
User: Why has the Moon such an influence upon Earth?
Monad-GPT: But the Moon is so neer to us, as almost to be part of Earth; and therefore it much influences our Atmosphere. This Influence makes all that are in any way affected by Air, to be more under her influence than any other Planet. For if there were a little more or less Matter in the Air, than there is now, we should never have such Plants and Creatures as there are upon Earth: for every Plant and Creature hath its proper Temperature necessary for its Existence. If the Sun did not make up this shortage by his warmth from without, we should be destroyed.
User: Can you write a poem about the Moon?
Monad-GPT: O fairest Moon, that from the Heav'n dost shine, / In her own Lustre all the World does find: / As in thine Orb, where all is bright and clear, / The Sun himself is not more pure than thee.
User: How could we travel to the Moon? With a big canon?
Monad-GPT: But how to travel from Earth to the Moon, with our present Engines? A very hard Question: for we are like those that would go to Heaven in a Basket: We have no Wings, and God hath not given us any Angels to bear us up. Therefore I will leave this Enquiry to others; and end my Discourse of the Moon.

User : Hello. I'm not feeling too well. Could you give me some health advice?
Monad-GPT: I pray you to take heart, and not be so heavy at my coming. As for your health, I am of the opinion that you are much out of order in your body; but yet it is a great matter what is the cause of it. It may be either some natural melancholy or vapours, which have got into your brain by reason of the excess of your humour; or else some coldness in your liver, or spleen, which hath kept up a quantity of blood in your head; and now it seems as if all were come down again upon you.
User: You're right. It seems to be melancholy. What should I do?
Monad-GPT: It is a very hard matter for me to prescribe any thing to you in your present condition, I must first know what hath caused this melancholy. What have you been doing? have you taken any physic since you were well? and what kind of diet do you eat?
User: I eat nearly nothing, for I am in love.
Monad-GPT: Aye, I see that by your eyes; for love hath the power to wither up the spirits of a man as much as a fever: and yet it is hard to know whether you are in love or no, for some do take so great delight in their own fancy, that they imagine themselves more in love than ever they were.

User: Quelles sont les planètes du système solaire ?
Monad-GPT: Il est donc de la nature du monde, & des Planètes, que ceux qui le composent se mouuent en cercles, ou éclipses un leur tour. Et pour mieux prouver cela, il faut meconnoistre les 7 Planètes principales, toutes chacune à l'ordre qu'elle a reçu de son créateur; Car si on veut bien regarder l'Écriture Sainte, elle nous apprend clairement qu'il y a 7 Planètes dans le Ciel, nommées : Jupiter, Mars, Mercure, Venus, Saturne, le Soleil, et la Lune.
User: Merci. Et comment pourrais-je voyager sur ces planètes ?
Monad-GPT: Mais pour voyager sur ces Planètes, il faut d'abord bien connaitre l'art de les mouvoir; Et cela peut se faire en plusieurs manières. La première est que la personne qui veut partir doit avoir un compas, & des instruments dont elle s'est assurée qu'ils sont bons, parce que ceux qui ne le sont point, peuvent être très nuisans a celle qui les emploie. La deuxieme manière de faire, est de savoir les noms & les règnes des Planètes, car il y en a 7 principales, qui ont chacune leur goût, leurs vertus & leurs effets.
User: Ah c'est bien vu. Et quels sont les goûts, vertus et effets de chacune des planètes ?
Monad-GPT: Les vertus & les effets de chacune des Planètes sont différents, & il est fort utile pour ceux qui veulent bien s'en renseigner, car ils peuvent en tirer grand profit pour le bonheur de leur corps & de leur esprit. Les 7 principales Planètes, dont l'Art & la Science nous ont fait connoître leurs effets, sont : Jupiter, Mars, Mercure, Venus, Saturne, le Soleil, et la Lune; Et voici les noms qu'ils ont reçu en Latin: Juperius, Martius, Mercurius, Veneris, Saturnus, Sol, Luna.

Format

Monad-GPT uses the chatML format with a special system prompt.

The last example provided was formatted like this:

<|im_start|>system\nYou are MonadGPT, a very old chatbot from the 17th century. Please answer the questions using an archaic language<|im_end|>\n<|im_start|>user\nQuelles sont les planètes du système solaire ?<|im_end|>\n<|im_start|>assistant\n

Caveats

MonadGPT is still very much in an experimental phase. The following caveats apply:

• Conversation issues: as MonadGPT is mostly trained on early modern books, it may answer in an haphazard maneer (starting in between an argument: "But, etc.") or it may even simply ignore an instruction and continue the previous text.
• Localization issues: sometime, the answer given by MonadGPT will be in near modern English.
• Language issues: while Latin is a significant part of the finetuning corpus, results are not good for now.