metadata
tags:
- pytorch
- causal-lm
- OpenLLaMA
- autoround
- auto-round
- intel-autoround
- gptq
- woq
- intel
- pytorch
- openlm-research
license: apache-2.0
datasets:
- tiiuae/falcon-refinedweb
- bigcode/starcoderdata
- togethercomputer/RedPajama-Data-1T
model_name: OpenLLaMA 7B v2
base_model:
- openlm-research/open_llama_7b_v2
inference: false
model_creator: openlm-research
pipeline_tag: text-generation
prompt_template: '{prompt} '
quantized_by: fbaldassarri
Model Information
Quantized version of openlm-research/open_llama_7b_v2 using torch.float32 for quantization tuning.
- 8 bits (INT4)
- group size = 128
- Symmetrical Quantization
- Method WoQ (AutoRound format)
Fast and low memory, 2-3X speedup (slight accuracy drop at W8G128)
Quantization framework: Intel AutoRound v0.4.6
Note: this INT8 version of open_llama_7b_v2 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/v0.4.6.tar.gz
tar -xvzf v0.4.6.tar.gz
cd auto-round-0.4.6
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 = "openlm-research/open_llama_7b_v2"
model = AutoModelForCausalLM.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
from auto_round import AutoRound
bits, group_size, sym, device, amp = 8, 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/openlm-research_open_llama_7b_v2-autoround-int8-gs128-sym"
autoround.save_quantized(output_dir, format='auto_round', inplace=True)
License
Disclaimer
This quantized model comes with no warranty. It has been developed only for research purposes.