How to use Varosa/llama-model-quantized with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Varosa/llama-model-quantized", dtype="auto")
What is a pickle import?
