{ "architectures": [ "Owlv2ForObjectDetection" ], "dtype": "float32", "initializer_factor": 1.0, "logit_scale_init_value": 2.6592, "model_type": "owlv2", "projection_dim": 64, "text_config": { "attention_dropout": 0.0, "bos_token_id": 49406, "dtype": "float32", "eos_token_id": 49407, "hidden_act": "quick_gelu", "hidden_size": 64, "initializer_factor": 1.0, "initializer_range": 0.02, "intermediate_size": 128, "layer_norm_eps": 1e-05, "max_position_embeddings": 16, "model_type": "owlv2_text_model", "num_attention_heads": 4, "num_hidden_layers": 2, "pad_token_id": 0, "vocab_size": 49408 }, "transformers_version": "5.22.0.dev0", "vision_config": { "attention_dropout": 0.0, "dtype": "float32", "hidden_act": "quick_gelu", "hidden_size": 64, "image_size": 960, "initializer_factor": 1.0, "initializer_range": 0.02, "intermediate_size": 128, "layer_norm_eps": 1e-05, "model_type": "owlv2_vision_model", "num_attention_heads": 4, "num_channels": 3, "num_hidden_layers": 2, "patch_size": 16 } }