| attention_dropout: 0.0 | |
| bos_token_id: 12 | |
| eos_token_id: 13 | |
| hidden_size: 768 | |
| intermediate_size: 768 | |
| learning_rate: 0.0001 | |
| max_epochs: 7 | |
| max_position_embeddings: 512 | |
| num_attention_heads: 8 | |
| num_hidden_layers: 7 | |
| num_labels: 105 | |
| pad_token_id: 0 | |
| steps_per_epoch: 375000 | |
| vocab_size: 591 | |
| warmup_epochs: 1 | |