| license: apache-2.0 | |
| Genearted from https://github.com/yhyu13/AutoGPTQ.git branch cuda_dev | |
| Original weight: https://huggingface.co/tiiuae/falcon-7b-instruct | |
| Note, autogptq does not generate 128 group size successfully when evaluating, at this moment. So the group size is 64 |