| { | |
| "_name_or_path": "./LongVILA-R1-7B/vision_tower", | |
| "architectures": [ | |
| "SiglipVisionModel" | |
| ], | |
| "attention_dropout": 0.0, | |
| "hidden_act": "gelu_pytorch_tanh", | |
| "hidden_size": 1152, | |
| "image_size": 448, | |
| "intermediate_size": 4304, | |
| "layer_norm_eps": 1e-06, | |
| "model_type": "siglip_vision_model", | |
| "num_attention_heads": 16, | |
| "num_channels": 3, | |
| "num_hidden_layers": 27, | |
| "num_image_tokens": 256, | |
| "patch_size": 14, | |
| "projection_dim": 2048, | |
| "projector_hidden_act": "gelu_fast", | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.46.2", | |
| "vision_use_head": false, | |
| "_attn_implementation": "flash_attention_2" | |
| } | |