Comments (2)
Thanks so much for your like and support for this project! ❤️
Ooba made his own fork for stability, but this does not support the latest GPTQ methods; specifically, --act-order. If a user wants to use --act-order, they have to link qwopqwop's latest GPTQ-for-LlaMa in instead.
Yes, I also experimented the two new features --act-order
and --true-sequential
on GPTJ model and found out it can truly bridge the gap between the quantized model and the origin one.
- Thus, I add the option of using
--act-order
toBaseQuantizeConfig
and renamed todesc_act
; - And for
--true-sequential
it has been add in this project as a non-optional feature, as you can see in this example
There is my roughly schedule and todo list of features on this project in the rest of days of this month, but it might be changed:
- I would spend ~2 hours per day on this project in my spare time
- Test on and add in more CausalLMs in transformers
- Faster version CUDA kernels in earlier version of gptq-for-llama cuda branch would be added back, but this may not compatible with
desc_act
as described in here - Triton integration, but I'm really new to triton so I've turned to qwopqwop200 for some advices and help, current and future discussions will all be in here. Also it would be wonderful if you have some more tips about development with triton, and even better to directly contribute to this project. 😄
And in particular I think it could be really good if oobabooga implemented your code in text-generation-webui.
AutoGPTQ is still in a very early stage and many apis might be vastly changed in the future, but I will pull a request to text-generation-webui as soon as AutoGPTQ becomes stable.
from autogptq.
If you don't use groupsize I think you can use desc_act and true sequential.
I think it's better to have those and not have group size but that's just me.
from autogptq.
Related Issues (20)
- [FEATURE] ADD Support DBRX HOT 16
- zeros remain zero?
- [FEATURE] ADD Jamba Support
- [BUG] Can not save quantized model to disk: "you shouldn't move a model that is dispatched using accelerate hooks." HOT 1
- Error when trying to quantize the JAIS model. HOT 10
- [BUG] GPTQ Kernels dont work with PEFT
- Error when quantizing mixtral 8x7b model. "ZeroDivisionError: float division by zero " HOT 1
- TypeError: forward() missing 1 required positional argument: 'hidden_states'[BUG] ? HOT 3
- [BUG]GPTQ QWEN-72B-Chat HOT 4
- gptq 4bit avg loss is large HOT 3
- export mistral8x7b error
- [BUG] Llama 3 8B Instruct - `no_inject_fused_attention` must be true or else errors out HOT 8
- Why doesn't AutoGPTQ quantize lm_head layer? HOT 5
- What magnitude of avg loss indicates a relatively good result for a quantization model HOT 6
- Why LLaMA3-8B after GPTQ test in wikitext2 so bad? HOT 8
- [PR Ready for Review] [FEATURE] Extend Support for Phi-3
- [FEATURE] Backport vllm expanded Marlin kernel to autogptq. HOT 1
- [DEPRECATION] Discussion on Fused attention and QiGEN HOT 5
- Llama-3 8B Instruct quantized to 8 Bit spits out gibberish in transformers `model.generate()` but works fine in vLLM? HOT 5
- [BUG]safetensors_rust.SafetensorError: Error while deserializing header: MetadataIncompleteBuffer
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from autogptq.