TheBlokeAI

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


Dans AdventurousWinds 7B - GPTQ

Description

This repo contains GPTQ model files for PocketDoc Labs's Dans AdventurousWinds 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.

Repositories available

Prompt template: Adventure

[Genres: Science Fiction]
[Tags: humor, old school, sci fi]
[Mode: Adventure]
[Description: A puzzle about committing acts of financial skulduggery and exploiting ridiculous magical items.]
[Misc: Writing era: 1993]
[Intro]
It is the year 2045. You are a young man in his twenties living in New York City. Your father was an inventor who died when you were very small; your mother raised you alone for many years until she remarried. Now you live with your stepfather, but he doesn't care much for you and has never given you any money to help support yourself. You have no job and little hope of getting one because of your lack of experience. However, you do have some unusual abilities that could be put to good use if only you knew how...

> {prompt}

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 models in 4-bit.
Branch Bits GS Act Order Damp % GPTQ Dataset Seq Len Size ExLlama Desc
main 4 128 Yes 0.1 wikitext 32768 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 wikitext 32768 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 wikitext 32768 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 wikitext 32768 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 wikitext 32768 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 wikitext 32768 4.29 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/Dans-AdventurousWinds-7B-GPTQ in the "Download model" box.

To download from another branch, add :branchname to the end of the download name, eg TheBloke/Dans-AdventurousWinds-7B-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 Dans-AdventurousWinds-7B-GPTQ:

mkdir Dans-AdventurousWinds-7B-GPTQ
huggingface-cli download TheBloke/Dans-AdventurousWinds-7B-GPTQ --local-dir Dans-AdventurousWinds-7B-GPTQ --local-dir-use-symlinks False

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

mkdir Dans-AdventurousWinds-7B-GPTQ
huggingface-cli download TheBloke/Dans-AdventurousWinds-7B-GPTQ --revision gptq-4bit-32g-actorder_True --local-dir Dans-AdventurousWinds-7B-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 Huggingface 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 Dans-AdventurousWinds-7B-GPTQ
HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/Dans-AdventurousWinds-7B-GPTQ --local-dir Dans-AdventurousWinds-7B-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/Dans-AdventurousWinds-7B-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/Dans-AdventurousWinds-7B-GPTQ.
  • To download from a specific branch, enter for example TheBloke/Dans-AdventurousWinds-7B-GPTQ:gptq-4bit-32g-actorder_True
  • see Provided Files above for the list of branches for each option.
  1. Click Download.
  2. The model will start downloading. Once it's finished it will say "Done".
  3. In the top left, click the refresh icon next to Model.
  4. In the Model dropdown, choose the model you just downloaded: Dans-AdventurousWinds-7B-GPTQ
  5. The model will automatically load, and is now ready for use!
  6. 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.
  1. 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/Dans-AdventurousWinds-7B-GPTQ --port 3000 --quantize awq --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'''[Genres: Science Fiction]
[Tags: humor, old school, sci fi]
[Mode: Adventure]
[Description: A puzzle about committing acts of financial skulduggery and exploiting ridiculous magical items.]
[Misc: Writing era: 1993]
[Intro]
It is the year 2045. You are a young man in his twenties living in New York City. Your father was an inventor who died when you were very small; your mother raised you alone for many years until she remarried. Now you live with your stepfather, but he doesn't care much for you and has never given you any money to help support yourself. You have no job and little hope of getting one because of your lack of experience. However, you do have some unusual abilities that could be put to good use if only you knew how...

> {prompt}
'''

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/Dans-AdventurousWinds-7B-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'''[Genres: Science Fiction]
[Tags: humor, old school, sci fi]
[Mode: Adventure]
[Description: A puzzle about committing acts of financial skulduggery and exploiting ridiculous magical items.]
[Misc: Writing era: 1993]
[Intro]
It is the year 2045. You are a young man in his twenties living in New York City. Your father was an inventor who died when you were very small; your mother raised you alone for many years until she remarried. Now you live with your stepfather, but he doesn't care much for you and has never given you any money to help support yourself. You have no job and little hope of getting one because of your lack of experience. However, you do have some unusual abilities that could be put to good use if only you knew how...

> {prompt}
'''

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 AutoGPTQ, both via Transformers and using AutoGPTQ directly. They should also work with Occ4m's GPTQ-for-LLaMa fork.

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

Huggingface Text Generation Inference (TGI) is compatible with all GPTQ models.

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.

Special thanks to: Aemon Algiz.

Patreon special mentions: Pierre Kircher, Stanislav Ovsiannikov, Michael Levine, Eugene Pentland, Andrey, 준교 김, Randy H, Fred von Graf, Artur Olbinski, Caitlyn Gatomon, terasurfer, Jeff Scroggin, James Bentley, Vadim, Gabriel Puliatti, Harry Royden McLaughlin, Sean Connelly, Dan Guido, Edmond Seymore, Alicia Loh, subjectnull, AzureBlack, Manuel Alberto Morcote, Thomas Belote, Lone Striker, Chris Smitley, Vitor Caleffi, Johann-Peter Hartmann, Clay Pascal, biorpg, Brandon Frisco, sidney chen, transmissions 11, Pedro Madruga, jinyuan sun, Ajan Kanaga, Emad Mostaque, Trenton Dambrowitz, Jonathan Leane, Iucharbius, usrbinkat, vamX, George Stoitzev, Luke Pendergrass, theTransient, Olakabola, Swaroop Kallakuri, Cap'n Zoog, Brandon Phillips, Michael Dempsey, Nikolai Manek, danny, Matthew Berman, Gabriel Tamborski, alfie_i, Raymond Fosdick, Tom X Nguyen, Raven Klaugh, LangChain4j, Magnesian, Illia Dulskyi, David Ziegler, Mano Prime, Luis Javier Navarrete Lozano, Erik Bjäreholt, 阿明, Nathan Dryer, Alex, Rainer Wilmers, zynix, TL, Joseph William Delisle, John Villwock, Nathan LeClaire, Willem Michiel, Joguhyik, GodLy, OG, Alps Aficionado, Jeffrey Morgan, ReadyPlayerEmma, Tiffany J. Kim, Sebastain Graf, Spencer Kim, Michael Davis, webtim, Talal Aujan, knownsqashed, John Detwiler, Imad Khwaja, Deo Leter, Jerry Meng, Elijah Stavena, Rooh Singh, Pieter, SuperWojo, Alexandros Triantafyllidis, Stephen Murray, Ai Maven, ya boyyy, Enrico Ros, Ken Nordquist, Deep Realms, Nicholas, Spiking Neurons AB, Elle, Will Dee, Jack West, RoA, Luke @flexchar, Viktor Bowallius, Derek Yates, Subspace Studios, jjj, Toran Billups, Asp the Wyvern, Fen Risland, Ilya, NimbleBox.ai, Chadd, Nitin Borwankar, Emre, Mandus, Leonard Tan, Kalila, K, Trailburnt, S_X, Cory Kujawski

Thank you to all my generous patrons and donaters!

And thank you again to a16z for their generous grant.

Original model card: PocketDoc Labs's Dans AdventurousWinds 7B

What is the model for?

This model is proficient in crafting text-based adventure games. It can both concise replies and more expansive, novel-like descriptions. The ability to alternate between these two response styles can be triggered by a distinct system message.

What's in the sausage?

This model was trained on Mistral-7b using a custom version of the floyd and cys datasets from the skein text adventure dataset. The dataset has been improved over what was used in my previous model "Dans-CreepingSenseOfDoom" and this time presented in the more traditional adventure mode format.

Prompt format and examples:

Format description:

Traditional adventure format with a description of the genre, tags, and other misc information at the top followed by an optional introductory message. User messages are prefixed with two new lines and "> ", model messages are one new line after the last user message or the description message if an introductory message is desired.

Examples of shorter responses:

[Genres: Science Fiction]
[Tags: humor, old school, sci fi]
[Mode: Adventure]
[Description: A puzzle about committing acts of financial skulduggery and exploiting ridiculous magical items.]
[Misc: Writing era: 1993]
[Intro]
It is the year 2045. You are a young man in his twenties living in New York City. Your father was an inventor who died when you were very small; your mother raised you alone for many years until she remarried. Now you live with your stepfather, but he doesn't care much for you and has never given you any money to help support yourself. You have no job and little hope of getting one because of your lack of experience. However, you do have some unusual abilities that could be put to good use if only you knew how...

> Look around
You are standing on the sidewalk outside your apartment building. The street is busy with traffic and people going about their daily business. Across the street from you stands a tall office tower where your stepfather works as an accountant. To your left is a convenience store where you can buy food and drinks.

> You go into the store
The doorbell rings as you enter the store. An elderly woman behind the counter looks up at you expectantly. "Hello," she says cheerfully, "what can I get for you?"

> You tell her you'll browse
"All right then," she replies brightly, "just take your time."

> You ask if she's hiring
"I'm sorry," she says sadly, "but we don't need anyone new just now."

> You leave without buying anything
As you walk out of the store, the bell jingles merrily.

> You go see your stepfather
When you arrive at work, your stepfather greets you warmly. He invites you into his office and offers you a seat. "So," he begins eagerly, "how would you like to make some extra cash? There's this guy named Mr. Jones who owes my company thousands of dollars. If we could somehow convince him to pay us back, it would really help our bottom line."

He pauses expectantly, waiting for your response.

> You ask him whats involved
"Well," he continues excitedly, "it turns out that Mr. Jones has a very valuable antique watch that he keeps locked away in a safe deposit box at the bank. We think that if we could somehow break into the box and steal the watch, we could sell it for enough money to cover his debt."

Your heart races as you realize the potential danger of such a scheme. But before you can say anything, your stepfather cuts off your thoughts by continuing, "Of course, there's always the chance that something could go wrong and we might end up losing everything..."

Examples of longer novel like responses:

[Genres: grimdark fantasy] [Tags: Fantasy, Grimdark, Serious, Villain Protagonist, War] [Description: This story is definitely more "story" than "game", however there are two distinct paths that can be followed, each with their own pitfalls and additional choices. This story can get long, and assuming you're surviving, years do pass in it and you and your environment will change.] [First message intro] [Mode: Interactive Storyteller]
You have been a soldier for as long as you remember. You don't know who you were before the war began, but you were someone else. Someone less angry. More forgiving. Perhaps even kinder.

It all started when you were very young. The King of this land was attacked by an army from another kingdom. His forces were outmatched, his city overrun. He had no choice but to flee into the forest, taking what few soldiers he could muster along with him.

As the king fled deeper into the woods, his army fell back, forming a protective wall around them. They fought valiantly against the invaders, giving the king time to escape, but they eventually succumbed to overwhelming numbers. When the fighting finally ended, only one man remained alive; you.

The king found you hiding behind some bushes, covered in blood and dirt. He took pity on you and offered you a place at court if you would swear allegiance to him. Without hesitation, you accepted. And so began your journey down the path of darkness...

> You meet with the king to discuss the plans
A week has passed since the battle and you have just met with the king to discuss future plans. It seems like things are finally starting to settle down after such chaos. However, there is still much work to be done before we can truly consider ourselves safe again.

The first order of business is rebuilding our army. We need men willing to fight for us once more. Men who understand what needs to be done and aren't afraid to die trying. I believe I have found such warriors among those loyal subjects who survived the initial attack. If we train them properly, then perhaps they might prove useful against our enemies next time around.

Next, we must find ways to improve our defenses. There are many areas where improvements could be made. For instance, we should construct better walls around our cities. These walls must not only keep intruders out, but also allow us easy access inside should anyone try to breach them. In addition, we must strengthen our gates so that no matter how strong our enemy becomes, he cannot break through without great effort. Finally, we must fortify our castles and other strategic locations throughout the country. By doing these things, we will ensure that nobody ever attacks us again!

> You ask what he desires your involvement to be
After hearing about his plans, you feel compelled to help him achieve victory. After all, you owe everything to him. But before agreeing to anything, you want to make sure that he understands exactly what you expect from him. So you decide to approach him directly.

"Your Majesty," you begin, "I am honored beyond measure to serve under your command. However, before committing myself fully to this cause, I wish to clarify certain matters."

He nods slowly. "Very well. What would you like me to explain?"

"Well," you say, "first off, I want to know why you chose me above everyone else? Why did you pick me specifically?"

He smiles softly. "Because I saw potential within you. Potential which others lacked. Potentials which could prove valuable during times of crisis. Therefore, I felt confident enough to entrust my fate to you."

"But what makes you think I possess such abilities?" you ask curiously.

"Oh, nothing special really," he replies casually. "Just raw determination combined with intelligence and resourcefulness. Those qualities alone are enough to carry us forward toward victory!"

Some quick and dirty training details:

  • Built with Axolotl
  • Sequence length: 4096
  • # of epochs: 3
  • Training time: 1 hour
  • Hardware: 1x RTX 3090
  • Training type: QLoRA
  • PEFT R/A: 32/32

Credits:

Skein Text Adventure Data:

Thank you to the Kobold AI community for curating the Skein dataset, which is pivotal to this model's capabilities.

Downloads last month
22
Safetensors
Model size
1.2B params
Tensor type
I32
·
FP16
·
Inference Examples
Inference API (serverless) has been turned off for this model.

Model tree for TheBloke/Dans-AdventurousWinds-7B-GPTQ

Quantized
(3)
this model