{ "architectures": [ "ChineseCLIPDualTextModel" ], "initializer_factor": 1.0, "initializer_range": 0.02, "logit_scale_init_value": 2.6592, "model_type": "chinese_clip", "projection_dim": 512, "text_config": { "_attn_implementation_autoset": true, "architectures": [ "ChineseCLIPTextModel" ], "attention_probs_dropout_prob": 0.1, "bos_token_id": 0, "directionality": "bidi", "eos_token_id": 2, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 768, "initializer_factor": 1.0, "initializer_range": 0.02, "intermediate_size": 3072, "layer_norm_eps": 1e-12, "max_position_embeddings": 512, "model_type": "chinese_clip_text_model", "num_attention_heads": 12, "num_hidden_layers": 12, "output_past": true, "pooler_fc_size": 768, "pooler_num_attention_heads": 12, "pooler_num_fc_layers": 3, "pooler_size_per_head": 128, "pooler_type": "first_token_transform", "position_embedding_type": "absolute", "torch_dtype": "float32", "type_vocab_size": 2, "use_cache": true, "vocab_size": 21128 }, "torch_dtype": "float32", "transformers_version": "4.51.3", "vision_config": { "attention_dropout": 0.0, "dropout": 0.0, "hidden_act": "quick_gelu", "hidden_size": 768, "image_size": 224, "initializer_factor": 1.0, "initializer_range": 0.02, "intermediate_size": 3072, "layer_norm_eps": 1e-05, "model_type": "chinese_clip_vision_model", "num_attention_heads": 12, "num_channels": 3, "num_hidden_layers": 12, "patch_size": 16, "projection_dim": 512, "torch_dtype": "float32" } }