News for hackers. Stuff that matters.
Re:InternLM – new open source 7B LLM by ipsum2 on Thursday July 06, 2023 @04:01AM (#36612703)
I believe it's because the model architecture isn't added to Huggingface transformer library, so it needs to eval some python code (i.e. load a pickle) to create the PyTorch model. Have not noticed it to be specific to models from China, almost all lesser known models have to do this.

There are no comments on this comment yet.