fbaldassarri's picture
Initial Upload
c376c40 verified
metadata
language:
  - en
tags:
  - pytorch
  - causal-lm
  - pythia
  - autoround
  - intel
  - auto-awq
  - autoawq
  - awq
  - woq
license: apache-2.0
model_name: Pythia 410m deduped
base_model: EleutherAI/pythia-410m-deduped
inference: false
model_creator: EleutherAI
datasets:
  - EleutherAI/pile
pipeline_tag: text-generation
prompt_template: '{prompt} '
quantized_by: fbaldassarri

Model Information

Quantized version of EleutherAI/pythia-410m-deduped using torch.float32 for quantization tuning.

  • 4 bits (INT4)
  • group size = 64
  • Symmetrical Quantization
  • Method AutoAWQ format

Quantization framework: Intel AutoRound v0.5.1

Note: this INT4 version of pythia-410m-deduped has been quantized to run inference through CPU.

Replication Recipe

Step 1 Install Requirements

I suggest to install requirements into a dedicated python-virtualenv or a conda enviroment.

wget https://github.com/intel/auto-round/archive/refs/tags/v0.5.1.tar.gz
tar -xvzf v0.5.1.tar.gz
cd auto-round-0.5.1
pip install -r requirements-cpu.txt --upgrade

Step 2 Build Intel AutoRound wheel from sources

pip install -vvv --no-build-isolation -e .[cpu]

Step 3 Script for Quantization

  from transformers import AutoModelForCausalLM, AutoTokenizer
  model_name = "EleutherAI/pythia-410m-deduped"
  model = AutoModelForCausalLM.from_pretrained(model_name)
  tokenizer = AutoTokenizer.from_pretrained(model_name)
  from auto_round import AutoRound
  bits, group_size, sym, device, amp = 4, 64, True, 'cpu', False
  autoround = AutoRound(model, tokenizer, nsamples=128, iters=200, seqlen=512, batch_size=4, bits=bits, group_size=group_size, sym=sym, device=device, amp=amp)
  autoround.quantize()
  output_dir = "./AutoRound/EleutherAI_pythia-410m-deduped-autoawq-int4-gs64-sym"
  auto

## License

[Apache 2.0 License](https://choosealicense.com/licenses/apache-2.0/)

## Disclaimer

This quantized model comes with no warrenty. It has been developed only for research purposes.