--- language: - en tags: - pytorch - causal-lm - pythia - autoround - intel - intel-autoround - awq - autoawq - auto-awq - woq license: apache-2.0 model_name: Pythia 6.9b base_model: EleutherAI/pythia-6.9b 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-6.9b](https://huggingface.co/EleutherAI/pythia-6.9b) using torch.float32 for quantization tuning. - 4 bits (INT4) - group size = 128 - Asymmetrical Quantization - Method AutoAWQ Quantization framework: [Intel AutoRound](https://github.com/intel/auto-round) v0.4.2 Note: this INT4 version of pythia-6.9b 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. ``` python -m pip install <package> --upgrade ``` - accelerate==1.2.0 - autoawq==0.2.7.post3 - auto_gptq==0.7.1 - neural_compressor==3.1.1 - torch==2.4.1+cpu - torchaudio==2.4.1+cpu - torchvision==0.19.1+cpu - transformers==4.47.0 ### Step 2 Build Intel Autoround wheel from sources ``` python -m pip install git+https://github.com/intel/auto-round.git ``` ### Step 3 Script for Quantization ``` from transformers import AutoModelForCausalLM, AutoTokenizer model_name = "EleutherAI/pythia-6.9b" model = AutoModelForCausalLM.from_pretrained(model_name) tokenizer = AutoTokenizer.from_pretrained(model_name) from auto_round import AutoRound bits, group_size, sym, device, amp = 4, 128, False, '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-6.9b-autoawq-int4-gs128-asym" autoround.save_quantized(output_dir, format='auto_awq', inplace=True) ``` ## 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.