{ "architectures": [ "UdopForConditionalGeneration" ], "d_ff": 4096, "d_kv": 64, "d_model": 1024, "decoder_start_token_id": 0, "dense_act_fn": "relu", "dropout_rate": 0.1, "eos_token_id": 1, "feed_forward_proj": "relu", "image_size": 224, "initializer_factor": 1.0, "is_encoder_decoder": true, "is_gated_act": false, "layer_norm_epsilon": 1e-06, "max_2d_position_embeddings": 1024, "model_type": "udop", "num_channels": 3, "num_decoder_layers": 24, "num_heads": 16, "num_layers": 24, "pad_token_id": 0, "patch_size": 16, "relative_attention_max_distance": 128, "relative_attention_num_buckets": 32, "relative_bias_args": [ { "type": "1d" }, { "type": "horizontal" }, { "type": "vertical" } ], "torch_dtype": "float32", "transformers_version": "4.29.0.dev0", "use_cache": true, "vocab_size": 33201 }