Open
Description
I've converted gguf files with Pytorch files from SparseLLM/ReluLLaMA-7B and predictor files from PowerInfer/ReluLLaMA-7B-Predictor https://huggingface.co/PowerInfer/ReluLLaMA-7B-Predictor/tree/main. Based on my experiments of running gguf files that I've created based on README specifications, I found that model generates strange, non logical output text. However, when I run PowerInfer/ReluLLaMA-7B-PowerInfer-GGUF model https://huggingface.co/PowerInfer/ReluLLaMA-7B-PowerInfer-GGUF such issue doesn't arise. Could you give me an insight for this problem? Below are captured files for output text based on GGUF files that I've generated