| { |
| "architectures": [ |
| "SiglipModel" |
| ], |
| "initializer_factor": 1.0, |
| "model_type": "siglip", |
| "text_config": { |
| "attention_dropout": 0.0, |
| "hidden_act": "gelu_pytorch_tanh", |
| "hidden_size": 1152, |
| "intermediate_size": 4304, |
| "layer_norm_eps": 1e-06, |
| "max_position_embeddings": 64, |
| "model_type": "siglip_text_model", |
| "num_attention_heads": 16, |
| "num_hidden_layers": 27, |
| "projection_size": 1152, |
| "vocab_size": 32000 |
| }, |
| "torch_dtype": "float32", |
| "transformers_version": "4.52.4", |
| "vision_config": { |
| "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, |
| "patch_size": 14 |
| } |
| } |
|
|
