Inclusion of language-table dataset
#4
by
tsaditya
- opened
Hello team, Thanks for the great work!
Wondering if language-table dataset has really been used in the training of this model. The paper's Appendix-A Data Mixture details says so but I cannot find the norm_stats for language-table in the config.json of the model. (all the other datasets exist except language-table)
Is it expected?