Model
BERT Large TensorFlow checkpoint pretrained using AMP and LAMB optimizer
Use the NGC CLI to download:
Copied!
1 Version
19.03.1_amp_optim-lambSelected09/22/2022 5:34 PM UTC5.03 GBAccuracy: 00 EpochsBatch Size: 0GPU: V100 Copied!
19.03.1_amp_optim-lambSelected
09/22/2022 5:34 PM UTC5.03 GBAccuracy: 00 EpochsBatch Size: 0GPU: V100
Copied!
architecture
| Key | Value |
|---|---|
| purpose | Pretraining |
| type | Large |
performance
| Key | Value |
|---|---|
| training_loss | 1.57? |
training
| Key | Value |
|---|---|
| optimizer | LAMB |
| training_precision | AMP |
| dataset | Wikipedia;BookCorpus |