Orca Mini v3 70B - GPTQ
- Model creator: Pankaj Mathur
- Original model: Orca Mini v3 70B
Description
This repo contains GPTQ model files for Pankaj Mathur's Orca Mini v3 70B.
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
- GPTQ models for GPU inference, with multiple quantisation parameter options.
- 2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference
- Pankaj Mathur's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions
Prompt template: Orca-Hashes
### System:
{system_message}
### User:
{prompt}
### Assistant:
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.
All GPTQ files are made with AutoGPTQ.
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 issues with models that use Act Order plus Group Size. - 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 dataset used for quantisation. Using a dataset more appropriate to the model's training can improve quantisation accuracy. Note that the GPTQ 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 | None | Yes | 0.1 | wikitext | 4096 | 35.33 GB | Yes | Most compatible option. Good inference speed in AutoGPTQ and GPTQ-for-LLaMa. Lower inference quality than other options. |
gptq-4bit-32g-actorder_True | 4 | 32 | Yes | 0.1 | wikitext | 4096 | 40.66 GB | Yes | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. Poor AutoGPTQ CUDA speed. |
gptq-4bit-64g-actorder_True | 4 | 64 | Yes | 0.1 | wikitext | 4096 | 37.99 GB | Yes | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. |
gptq-4bit-128g-actorder_True | 4 | 128 | Yes | 0.1 | wikitext | 4096 | 36.65 GB | Yes | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. |
gptq-3bit--1g-actorder_True | 3 | None | Yes | 0.1 | wikitext | 4096 | 26.77 GB | No | 3-bit, with Act Order and no group size. Lowest possible VRAM requirements. May be lower quality than 3-bit 128g. |
gptq-3bit-128g-actorder_True | 3 | 128 | Yes | 0.1 | wikitext | 4096 | 28.03 GB | No | 3-bit, with group size 128g and act-order. Higher quality than 128g-False but poor AutoGPTQ CUDA speed. |
How to download from branches
- In text-generation-webui, you can add
:branch
to the end of the download name, egTheBloke/orca_mini_v3_70B-GPTQ:gptq-4bit-32g-actorder_True
- With Git, you can clone a branch with:
git clone --single-branch --branch gptq-4bit-32g-actorder_True https://huggingface.co/TheBloke/orca_mini_v3_70B-GPTQ
- In Python Transformers code, the branch is the
revision
parameter; see below.
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 know how to make a manual install.
- Click the Model tab.
- Under Download custom model or LoRA, enter
TheBloke/orca_mini_v3_70B-GPTQ
.
- To download from a specific branch, enter for example
TheBloke/orca_mini_v3_70B-GPTQ:gptq-4bit-32g-actorder_True
- see Provided Files above for the list of branches for each option.
- Click Download.
- The model will start downloading. Once it's finished it will say "Done"
- In the top left, click the refresh icon next to Model.
- In the Model dropdown, choose the model you just downloaded:
orca_mini_v3_70B-GPTQ
- The model will automatically load, and is now ready for use!
- 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 set GPTQ parameters any more. These are set automatically from the file
quantize_config.json
.
- Once you're ready, click the Text Generation tab and enter a prompt to get started!
How to use this GPTQ model from Python code
First make sure you have AutoGPTQ 0.3.1 or later installed:
pip3 install auto-gptq
If you have problems installing AutoGPTQ, please build from source instead:
pip3 uninstall -y auto-gptq
git clone https://github.com/PanQiWei/AutoGPTQ
cd AutoGPTQ
pip3 install .
Then try the following example code:
from transformers import AutoTokenizer, pipeline, logging
from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
model_name_or_path = "TheBloke/orca_mini_v3_70B-GPTQ"
use_triton = False
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
use_safetensors=True,
trust_remote_code=False,
device="cuda:0",
use_triton=use_triton,
quantize_config=None)
"""
# To download from a specific branch, use the revision parameter, as in this example:
# Note that `revision` requires AutoGPTQ 0.3.1 or later!
model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
revision="gptq-4bit-32g-actorder_True",
use_safetensors=True,
trust_remote_code=False,
device="cuda:0",
quantize_config=None)
"""
prompt = "Tell me about AI"
prompt_template=f'''### System:
{system_message}
### User:
{prompt}
### 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, max_new_tokens=512)
print(tokenizer.decode(output[0]))
# Inference can also be done using transformers' pipeline
# Prevent printing spurious transformers error when using pipeline with AutoGPTQ
logging.set_verbosity(logging.CRITICAL)
print("*** Pipeline:")
pipe = pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
max_new_tokens=512,
temperature=0.7,
top_p=0.95,
repetition_penalty=1.15
)
print(pipe(prompt_template)[0]['generated_text'])
Compatibility
The files provided will work with AutoGPTQ (CUDA and Triton modes), GPTQ-for-LLaMa (only CUDA has been tested), and Occ4m's GPTQ-for-LLaMa fork.
ExLlama works with Llama models in 4-bit. Please see the Provided Files table above for per-file compatibility.
Discord
For further support, and discussions on these models and AI in general, join us at:
Thanks, and how to contribute.
Thanks to the chirper.ai team!
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: Ajan Kanaga, David Ziegler, Raymond Fosdick, SuperWojo, Sam, webtim, Steven Wood, knownsqashed, Tony Hughes, Junyu Yang, J, Olakabola, Dan Guido, Stephen Murray, John Villwock, vamX, William Sang, Sean Connelly, LangChain4j, Olusegun Samson, Fen Risland, Derek Yates, Karl Bernard, transmissions 11, Trenton Dambrowitz, Pieter, Preetika Verma, Swaroop Kallakuri, Andrey, Slarti, Jonathan Leane, Michael Levine, Kalila, Joseph William Delisle, Rishabh Srivastava, Deo Leter, Luke Pendergrass, Spencer Kim, Geoffrey Montalvo, Thomas Belote, Jeffrey Morgan, Mandus, ya boyyy, Matthew Berman, Magnesian, Ai Maven, senxiiz, Alps Aficionado, Luke @flexchar, Raven Klaugh, Imad Khwaja, Gabriel Puliatti, Johann-Peter Hartmann, usrbinkat, Spiking Neurons AB, Artur Olbinski, chris gileta, danny, Willem Michiel, WelcomeToTheClub, Deep Realms, alfie_i, Dave, Leonard Tan, NimbleBox.ai, Randy H, Daniel P. Andersen, Pyrater, Will Dee, Elle, Space Cruiser, Gabriel Tamborski, Asp the Wyvern, Illia Dulskyi, Nikolai Manek, Sid, Brandon Frisco, Nathan LeClaire, Edmond Seymore, Enrico Ros, Pedro Madruga, Eugene Pentland, John Detwiler, Mano Prime, Stanislav Ovsiannikov, Alex, Vitor Caleffi, K, biorpg, Michael Davis, Lone Striker, Pierre Kircher, theTransient, Fred von Graf, Sebastain Graf, Vadim, Iucharbius, Clay Pascal, Chadd, Mesiah Bishop, terasurfer, Rainer Wilmers, Alexandros Triantafyllidis, Stefan Sabev, Talal Aujan, Cory Kujawski, Viktor Bowallius, subjectnull, ReadyPlayerEmma, zynix
Thank you to all my generous patrons and donaters!
Original model card: Pankaj Mathur's Orca Mini v3 70B
orca_mini_v3_70b
A Llama2-70b model trained on Orca Style datasets.
quantized versions
Big thanks to @TheBloke
license disclaimer:
This model is bound by the license & usage restrictions of the original Llama-2 model. And comes with no warranty or gurantees of any kind.
Evaluation
We evaluated orca_mini_v3_70b on a wide range of tasks using Language Model Evaluation Harness from EleutherAI.
Here are the results on metrics used by HuggingFaceH4 Open LLM Leaderboard
Task | Metric | Value | Stderr |
arc_challenge | acc_norm | 0.7098 | 0.0132 |
hellaswag | acc_norm | 0.8779 | 0.0032 |
mmlu | acc_norm | 0.6904 | 0.0351 |
truthfulqa_mc | mc2 | 0.6196 | 0.0151 |
Total Average | - | 0.722175 |
P.S. I am actively seeking sponsorship and partnership opportunities. If you're interested, please connect with me at www.linkedin.com/in/pankajam.
Example Usage
Here is the prompt format
### System:
You are an AI assistant that follows instruction extremely well. Help as much as you can.
### User:
Tell me about Orcas.
### Assistant:
Below shows a code example on how to use this model
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
tokenizer = AutoTokenizer.from_pretrained("psmathur/orca_mini_v3_70b")
model = AutoModelForCausalLM.from_pretrained(
"psmathur/orca_mini_v3_70b",
torch_dtype=torch.float16,
load_in_8bit=True,
low_cpu_mem_usage=True,
device_map="auto"
)
system_prompt = "### System:\nYou are an AI assistant that follows instruction extremely well. Help as much as you can.\n\n"
#generate text steps
instruction = "Tell me about Orcas."
prompt = f"{system_prompt}### User: {instruction}\n\n### Assistant:\n"
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
output = model.generate(**inputs, do_sample=True, top_p=0.95, top_k=0, max_new_tokens=4096)
print(tokenizer.decode(output[0], skip_special_tokens=True))
Limitations & Biases:
While this model aims for accuracy, it can occasionally produce inaccurate or misleading results.
Despite diligent efforts in refining the pretraining data, there remains a possibility for the generation of inappropriate, biased, or offensive content.
Exercise caution and cross-check information when necessary.
Citiation:
Please kindly cite using the following BibTeX:
@misc{orca_mini_v3_70b,
author = {Pankaj Mathur},
title = {orca_mini_v3_70b: An Orca Style Llama2-70b model},
year = {2023},
publisher = {HuggingFace},
journal = {HuggingFace repository},
howpublished = {\url{https://https://huggingface.co/psmathur/orca_mini_v3_70b},
}
@misc{mukherjee2023orca,
title={Orca: Progressive Learning from Complex Explanation Traces of GPT-4},
author={Subhabrata Mukherjee and Arindam Mitra and Ganesh Jawahar and Sahaj Agarwal and Hamid Palangi and Ahmed Awadallah},
year={2023},
eprint={2306.02707},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
@software{touvron2023llama2,
title={Llama 2: Open Foundation and Fine-Tuned Chat Models},
author={Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava,
Shruti Bhosale, Dan Bikel, Lukas Blecher, Cristian Canton Ferrer, Moya Chen, Guillem Cucurull, David Esiobu, Jude Fernandes, Jeremy Fu, Wenyin Fu, Brian Fuller,
Cynthia Gao, Vedanuj Goswami, Naman Goyal, Anthony Hartshorn, Saghar Hosseini, Rui Hou, Hakan Inan, Marcin Kardas, Viktor Kerkez Madian Khabsa, Isabel Kloumann,
Artem Korenev, Punit Singh Koura, Marie-Anne Lachaux, Thibaut Lavril, Jenya Lee, Diana Liskovich, Yinghai Lu, Yuning Mao, Xavier Martinet, Todor Mihaylov,
Pushkar Mishra, Igor Molybog, Yixin Nie, Andrew Poulton, Jeremy Reizenstein, Rashi Rungta, Kalyan Saladi, Alan Schelten, Ruan Silva, Eric Michael Smith,
Ranjan Subramanian, Xiaoqing Ellen Tan, Binh Tang, Ross Taylor, Adina Williams, Jian Xiang Kuan, Puxin Xu , Zheng Yan, Iliyan Zarov, Yuchen Zhang, Angela Fan,
Melanie Kambadur, Sharan Narang, Aurelien Rodriguez, Robert Stojnic, Sergey Edunov, Thomas Scialom},
year={2023}
}
- Downloads last month
- 12