Image Credits:<\/strong> Stability AI<\/p>\n<\/div>\n

Stability AI didn\u2019t say in the blog post whether the StableLM models suffer from the same limitations as others, namely a tendency to generate toxic responses to certain prompts and hallucinate (i.e. make up) facts. But given that The Pile contains profane, lewd and otherwise fairly abrasive language, it wouldn\u2019t be surprising if that were the case.<\/p>\n

This reporter tried to test the models on Hugging Face, which provides a front end to run them without having to configure the code from scratch. Unfortunately, I got an \u201cat capacity\u201d error every time, which might have to do with the size of the models \u2014 or their popularity.<\/p>\n

\u201cAs is typical for any pretrained large language model without additional fine-tuning and reinforcement learning, the responses a user gets might be of varying quality and might potentially include offensive language and views,\u201d Stability AI wrote in the repo for StableLM. \u201cThis is expected to be improved with scale, better data, community feedback and optimization.\u201d<\/p>\n