README.md · fbaldassarri/iGeniusAI_Italia-9B-Instruct-v0.1-autoawq-int4-gs128-sym at main

metadata

language:
  - it
tags:
  - pretrained
  - pytorch
  - causal-lm
  - autoround
  - intel-autoround
  - woq
  - awq
  - autoawq
  - auto-awq
  - intel
  - italia
  - italiano
  - italian
license: mit
license_link: https://huggingface.co/iGeniusAI/Italia-9B-Instruct-v0.1/blob/main/LICENSE
model_name: Italia 9B Instruct v0.1
base_model:
  - iGeniusAI/Italia-9B-Instruct-v0.1
inference: false
model_creator: iGeniusAI
pipeline_tag: text-generation
prompt_template: '{prompt} '
quantized_by: fbaldassarri

Model Information

Quantized version of iGeniusAI/Italia-9B-Instruct-v0.1 using torch.float32 for quantization tuning.

4 bits (INT4)
group size = 128
Symmetrical Quantization
Method AutoAWQ

Quantization framework: Intel AutoRound v0.4.3

Note: this INT4 version of Italia-9B-Instruct-v0.1 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/v0.4.3.tar.gz
tar -xvzf v0.4.3.tar.gz
cd auto-round-0.4.3
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, GPTNeoXModel
  model_name = "iGeniusAI/Italia-9B-Instruct-v0.1"
  model = GPTNeoXModel.from_pretrained(model_name, trust_remote_code=True)
  tokenizer = AutoTokenizer.from_pretrained(model_name)
  from auto_round import AutoRound
  bits, group_size, sym, device, amp = 4, 128, 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/iGeniusAI_Italia-9B-Instruct-v0.1-autoawq-int4-gs128-sym"
  autoround.save_quantized(output_dir, format='auto_awq', inplace=True)

Note: the GPTNeoXSdpaAttention class is deprecated in favor of simply modifying the config._attn_implementationattribute of the GPTNeoXAttention class. So this require transformers<4.48.

License

MIT

Disclaimer

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