Results for Optimizing OPT Model seems unusually high for 4 bits per tensor

Looking at the data for `4 bits per tensor` in the results table for optimizing OPT model, the values for `c4 Perplexity` and `wikitext2 Perplexity` seem unusually high compared to the other configurations.
Are these values correct or there is some typo?

Code link here: https://github.com/apple/coremltools/blob/c3ea4cf56fef1176417246c1b85363417f3e713d/docs-guides/source/opt-opt1_3.md?plain=1#L74


Open docs-guides/source/opt-opt1_3.md around line 74 and inspect the 4 bits per tensor row in the OPT results table. Compare the c4 Perplexity and wikitext2 Perplexity entries with the other configurations and verify whether they are intentional; done means the table is confirmed or the suspected typo is corrected.

Written by the indexing model from the issue text.