| { | |
| "seq_bytes": 2048, | |
| "hidden_size": 768, | |
| "num_hidden_layers": 12, | |
| "num_attention_heads": 12, | |
| "intermediate_size": 3072, | |
| "max_segments": 2, | |
| "hidden_dropout_prob": 0.1, | |
| "attention_probs_dropout_prob": 0.1 | |
| } | |
| { | |
| "seq_bytes": 2048, | |
| "hidden_size": 768, | |
| "num_hidden_layers": 12, | |
| "num_attention_heads": 12, | |
| "intermediate_size": 3072, | |
| "max_segments": 2, | |
| "hidden_dropout_prob": 0.1, | |
| "attention_probs_dropout_prob": 0.1 | |
| } | |