| {"vocab_size": 32, "dropout": 0.1, "hidden_size": 1024, "num_heads": 16, "num_layers": 24, "intermediate_size": 4096, "is_gelu_approx": false, "layer_norm_eps": 1e-05, "survival_prob": 1.0, "pad_id": 0, "num_conv_pos_embeddings": 128, "num_conv_pos_embedding_groups": 16, "filter_sizes": [512, 512, 512, 512, 512, 512, 512], "kernal_sizes": [10, 3, 3, 3, 3, 2, 2], "strides": [5, 2, 2, 2, 2, 2, 2], "conv_bias": true, "apply_spec_augment": true, "mask_time_prob": 0.05, "mask_time_length": 10, "attention_norm_type": "prenorm", "feature_extractor_norm_type": "layer", "is_robust": true} |