Katie Hunt
Results
3
issues of
Katie Hunt
Hi! I ran the training script on 130 million training instances and I got the following training speed: 1 V100 GPU, FP16 O2, ~14k tokens/sec, ~100 hours 8 V100 GPUs,...
Hey guys! Great work! I really appreciate it! After reading the code, I noticed that the training data is from 12/2015 to 11/2017, while the test data is from 03/2018...
Can you guys share the hyperparameters of different model sizes i.e. small, medium, and large? https://github.com/microsoft/DialoGPT/blob/75a4197188a1addf22c5eaea23f16d3b598635d7/LSP_train.py#L46-L82