Could you provide the training hyper paras

google-research / albert

ALBERT: A Lite BERT for Self-supervised Learning of Language Representations

Apache License 2.0

3.23k stars 571 forks source link

Could you provide the training hyper paras #235

Open juyiming opened 3 years ago

juyiming commented 3 years ago

hi,

Could you provide the training hyper paras of xxlarge version for MNLI, MRPC and SQUAD2.0? I only have a 1080Ti GPU for training, hyper params search is too difficult for me. thx.

RahulSChand commented 2 years ago

Hi @juyiming the hyper-parameters for all GLUE Tasks (which includes MNLI/MRPC) & SQUAD_V2.0 are available in the Appendix A4 section.