{ "architectures": [ "FastVlmForConditionalGeneration" ], "dtype": "bfloat16", "image_seq_length": 256, "image_token_index": 151646, "model_type": "fast_vlm", "multimodal_projector_bias": true, "projector_hidden_act": "gelu", "text_config": { "_name_or_path": "Qwen/Qwen2-7B", "architectures": [ "Qwen2ForCausalLM" ], "attention_dropout": 0.0, "bos_token_id": 151643, "dtype": "bfloat16", "eos_token_id": 151643, "hidden_act": "silu", "hidden_size": 3584, "initializer_range": 0.02, "intermediate_size": 18944, "layer_types": [ "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention" ], "max_position_embeddings": 131072, "max_window_layers": 28, "model_type": "qwen2", "num_attention_heads": 28, "num_hidden_layers": 28, "num_key_value_heads": 4, "rms_norm_eps": 1e-06, "rope_scaling": null, "rope_theta": 1000000.0, "sliding_window": null, "use_cache": true, "use_sliding_window": false, "vocab_size": 152128 }, "transformers_version": "4.56.0.dev0", "vision_config": { "_name_or_path": "timm/fastvit_mci3.apple_mclip2_dfndr2b", "architecture": "fastvit_mci3", "do_pooling": true, "global_pool": "avg", "hidden_size": 3072, "initializer_range": 0.02, "label_names": [], "model_args": { "inference_mode": true }, "model_type": "timm_wrapper", "num_classes": 0, "num_features": 3072, "pretrained_cfg": { "classifier": "head.fc", "crop_mode": "center", "crop_pct": 0.95, "custom_load": false, "first_conv": [ "stem.0.conv_kxk.0.conv", "stem.0.conv_scale.conv" ], "fixed_input_size": false, "input_size": [ 3, 256, 256 ], "interpolation": "bicubic", "license": "apple-amlr", "mean": [ 0.48145466, 0.4578275, 0.40821073 ], "pool_size": [ 8, 8 ], "std": [ 0.26862954, 0.26130258, 0.27577711 ], "tag": "apple_mclip2_dfndr2b" } }, "vision_feature_layer": -1, "vision_feature_select_strategy": "full" }