Official repository for "Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing". Your efficient and high-quality synthetic data generation pipeline!
While attempting to execute filtered_dataset = dataset['train'].filter(high_quality_filter), I encountered a KeyError: 'uuid'. Upon inspecting the code in exp/gen_dis.py, it appears that the uuid value is not assigned. Could this be an oversight, or am I misunderstanding something?
Dear authors,
Thanks for your excellent work!
While attempting to execute
filtered_dataset = dataset['train'].filter(high_quality_filter)
, I encountered aKeyError: 'uuid'
. Upon inspecting the code in exp/gen_dis.py, it appears that theuuid
value is not assigned. Could this be an oversight, or am I misunderstanding something?